Tony Pearson is a Master Inventor and Senior IT Architect for the IBM Storage product line at the
IBM Systems Client Experience Center in Tucson Arizona, and featured contributor
to IBM's developerWorks. In 2016, Tony celebrates his 30th year anniversary with IBM Storage. He is
author of the Inside System Storage series of books. This blog is for the open exchange of ideas relating to storage and storage networking hardware, software and services.
(Short URL for this blog: ibm.co/Pearson )
My books are available on Lulu.com! Order your copies today!
Safe Harbor Statement: The information on IBM products is intended to outline IBM's general product direction and it should not be relied on in making a purchasing decision. The information on the new products is for informational purposes only and may not be incorporated into any contract. The information on IBM products is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. The development, release, and timing of any features or functionality described for IBM products remains at IBM's sole discretion.
Tony Pearson is a an active participant in local, regional, and industry-specific interests, and does not receive any special payments to mention them on this blog.
Tony Pearson receives part of the revenue proceeds from sales of books he has authored listed in the side panel.
Tony Pearson is not a medical doctor, and this blog does not reference any IBM product or service that is intended for use in the diagnosis, treatment, cure, prevention or monitoring of a disease or medical condition, unless otherwise specified on individual posts.
Whew! I am glad that is over. The BarryB circus has left town, he has decided to [move on to other topics], and I am now to clean up the ["circus gold"] leftbehind. I would like to remind everyone that all of these discussions have been about the architecture,not the product. IBM will come out withits own version of a product based on Nextra later in 2008, which may be different than the product that XIV currentlysells to its customers.
RAID-X does not protect against double-drive failures as well as RAID-6, but it's very close
BarryB calls this the "Elephant in the room", that RAID-6 protects better against double-drive failures. I don't dispute that. He also credits me with the term "RAID-X", but I got this directly from the XIV guys. It turns out this was already a term used among academic research circles for [distributed RAID environments]. Meanwhile, Jon Toigo feels the term RAID-X sounds like a brand of bug spray in his post[XIV Architecture: What’s Not to Like?]Perhaps IBM can change this to RAID-5.99 instead.
If you measure risk of a second drive failing during the rebuild or re-replication process ofa first drive failure, you can measure the exposure by multiplying the amount of GB at risk by thenumber of hours that the second failure could occur, resulting in a unit of "GB-hours". Here Ilist best-case rebuild times, your mileage may vary depending on whether other workloads existon the system competing for resources. Notice that 8-disk configurations of RAID-10 and RAID-5for smaller FC disk are in the triple digits, and larger SATA disk in five digits, but that with RAID-X it is only single digits. That is orders of magnitude closer to the ideal.
For each RAID type, the risk is proportional to the square of the individual drive size.Double the drive size causes the risk to be four times greater.This is not the first time this has been discussed. In [Is RAID-5 Getting Old?], Ramskovquotes NetApp's response in Robin Harris' [NetApp Weighs In On Disks]:
...protecting online data only via RAID 5 today verges on professional malpractice.
As disks get older, RAID-6 will not be able to protect against 3-drive failures. A similar chartabove could show the risk to data after the second drive fails and both rebuilds are going on,compared to the risk of a third drive failure during this time. The RAID-X scheme protects muchbetter against 3-drive failures than RAID-6.
Nothing in the Nextra architecture prevents a RAID-6, Triple-copy, or other blob-level scheme
In much the same way that EMC Centera is RAID-5 based for its blobs, there is nothing in the Nextra architecturethat prevents taking additional steps to provide even better protection, using a RAID-6 scheme, making three copiesof the data instead of two copies, or something even more advanced. The current two-copy scheme for RAID-X is betterthan all the RAID-5 and RAID-10 systems out in the marketplace today.
Mirrored Cache won't protect against Cosmic rays, but ECC detection/correction does
BarryB incorrectly states that since some implementations of cache are non-mirrored, that this implies they are unprotected against Cosmic rays. Mirroring does not protect against bit-flips unless both copies arecompared for differences. Unfortunately, even if you compared them, the best you can do is detect theyare different, there is no way of knowing which version is correct.Mirroring cache is normally done to protect uncommitted writes. Reads in cacheare expendable copies of data already written to disk, so ECC detection/correction schemes are adequateprotection. ECC is like RAID for DRAM memory. A single bit-flip can be corrected, multiple bit-flipscan be detected. In the case of detection, the cache copy is discarded and read fresh again from disk.IBM DS8000, XIV and probably most other major vendor offerings use ECC of some kind. BarryB is correctthat some cheaper entry-level and midrange offerings from other vendors might cut corners in this area.I don't doubt BarryB's assertion that the ECC method used in the EMC products may be differently implemented than theECC in the IBM DS8000, but that doesn't mean the IBM DS8000's ECC implementation is flawed.
ECC protection is important for all RAID systems that perform rebuild, and even more importantthe larger the GB-hours listed in the table above.
XIV is designed for high-utilization, not less than 50 percent
I mentioned that the typical Linux, UNIX or Windows LUN is only 30-50 percent full, and perhaps BarryBthought I was referring to the typical "XIV customer". This average is for all disk storage systems connectedto these operating systems, based on IBM market research and analyst reports. The XIV is expected to run at much higher utilization rates, and offers features like "thin provisioning" and "differential snapshot" to make this simple to implement in practice.
Most often, disks don't fail without warning. Usually, they give out temporary errors first, and then fail permanently.The XIV architecture allows for pre-emptive self-repair, initiating the re-replication process after detecting temporary errors, rather than waiting for a complete drive failure.
I had mentioned that this process used "spare capacity, not spare drives" but I was notified that there are three spare drives per system to ensure that there is enough spare capacity, so I stand corrected.
New drives don't have to match the same speed/capacity as the new drives, so three to five years from now, whenit might be hard to find a matching 500GB SATA drive anymore, you won't have to.
No RAID scheme eliminates backups or Business Continuity Planning
The XIV supports both synchronous and asynchronous disk mirroring to remote locations. Backup software willbe able to backup data from the XIV to tape. A double drive failure would require a "recovery action", eitherfrom the disk mirror, or from tape, for the few GB of data that need to be recovered.
A third alternative is to allow end-users to receive backups of their own user-generated content. For example, I have over 15,000 photos uploaded over the past six years to Kodak Photo Gallery, which I use to share with my friends and family. For about $180 US dollars, they will cut DVDs containing all of my uploaded files and send them to me, so that I do not have to worry about Kodak losing my photos.In many cases, if a company or product fails to deliver on its promises, the most you will get is your money back, but for "free services" like HotMail, FreeDrive, FlickR and others, you didn't pay anything in the first place, andthey may point this limitation of liability in the "terms of service".
XIV can be used for databases and other online transaction processing
The XIV will have FCP and iSCSI interfaces, and systems can use these to store any kind of data you want. I mentionedthat the design was intended for large volumes of unstructured digital content, but there is nothing to prevent the use of other workloads. In today's Wall Street Journal article[To Get Back Into the Storage Game, IBM Calls In an Old Foe]:
Today, XIV's Nextra system is used by Bank Leumi, a large Israeli bank, and a few other customers for traditional data-storage tasks such as recording hundreds of transactions a minute.
BarryB, thanks for calling the truce. I look forward to talking about other topics myself. These past two weeks have been exhausting!
In my post yesterday [Spreading out the Re-Replication process], fellow blogger BarryB [aka The Storage Anarchist]raises some interesting points and questions in the comments section about the new IBM XIV Nextra architecture.I answer these below not just for the benefit of my friends at EMC, but also for my own colleagues within IBM,IBM Business Partners, Analysts and clients that might have similar questions.
If RAID 5/6 makes sense on every other platform, why not so on the Web 2.0 platform?
Your attempt to justify the expense of Mirrored vs. RAID 5 makes no sense to me. Buying two drives for every one drive's worth of usable capacity is expensive, even with SATA drives. Isn't that why you offer RAID 5 and RAID 6 on the storage arrays that you sell with SATA drives?
And if RAID 5/6 makes sense on every other platform, why not so on the (extremely cost-sensitive) Web 2.0 platform? Is faster rebuild really worth the cost of 40+% more spindles? Or is the overhead of RAID 6 really too much for those low-cost commodity servers to handle.
Let's take a look at various disk configurations, for example 3TB on 750GB SATA drives:
JBOD: 4 drives
JBOD here is industry slang for "Just a Bunch of Disks" and was invented as the term for "non-RAID".Each drive would be accessible independently, at native single-drive speed, with no data protection. Puttingfour drives in a single cabinet like this provides simplicity and convenience only over four separate drivesin their own enclosures.
RAID-10: 8 drives
RAID-10 is a combination of RAID-1 (mirroring) and RAID-0 (striping). In a 4x2 configuration, data is striped across disks 1-4,then these are mirrored across to disks 5-8. You get performance improvement and protection against a singledrive failure.
RAID-5: 5 drives
This would be a 4+P configuration, where there would be four drives' worth of data scattered across fivedrives. This gives you almost the same performance improvement as RAID-10, similar protection againstsingle drive failure, but with fewer drives per usable TB capacity.
RAID-6: 6 drives
This would be a 4+2P configuration, where the first P represents linear parity, and the second represents a diagonal parity. Similar in performance improvement as RAID-5, but protects against single and double drive failures, and still better than RAID-10 in terms of drives per TB usable capacity.
For all the RAID configurations, rebuild would require a spare drive, but often spares are shared among multiple RAID ranks, not dedicated to a single rank. To this end, you often have to have several spares per I/O loop, and a different set of spares for each kind of speed and capacity. If you had a mix of 15K/73GB, 10K/146GB, and 7200/500GB drives, then you would have three sets of spares to match.
In contrast, IBM XIV's innovative RAID-X approach doesn't requireany spare drives, just spare capacity on existing drives being used to hold data. The objects can be mirroredbetween any two types of drives, so no need to match one with another.
All of these RAID levels represent some trade-off between cost, protection and performance, and IBM offers each of theseon various disk systems platforms. Calculating parity is more complicated than just mirrored copies, but this can be done with specialized chips in cache memory to minimize performance impact.IBM generally recommends RAID-5 for high-performance FC disk, and RAID-6 for slower, large capacity SATA disk.
However, the questionassumes that the drive cost is a large portion of the overall "disk system" cost. It isn't. For example,Jon Toigo discusses the cost of EMC's new AX4 disk system in his post [National Storage Rip-Off Day]:
EMC is releasing its low end Clariion AX4 SAS/SATA array with 3TB capacity for $8600. It ships with four 750GB SATA drives (which you and I could buy at list for $239 per unit). So, if the disk drives cost $956 (presumably far less for EMC), that means buyers of the EMC wares are paying about $7700 for a tin case, a controller/backplane, and a 4Gbps iSCSI or FC connector. Hmm.
Dell is offering EMC’s AX4-5 with same configuration for $13,000 adding a 24/7 warranty.
(Note: I checked these numbers. $8599 is the list price that EMC has on its own website. External 750GB drivesavailable at my local Circuit City ranged from $189 to $329 list price. I could not find anything on Dell'sown website, but found [The Register] to confirm the $13,000 with 24x7 warranty figure.)
Disk capacity is a shrinking portion of the total cost of ownership (TCO). In addition to capacity, you are paying forcache, microcode and electronics of the system itself, along with software and services that are included in the mix,and your own storage administrators to deal with configuration and management. For more on this, see [XIV storage - Low Total Cost of Ownership].
EMC Centera has been doing this exact type of blob striping and protection since 2002
As I've noted before, there's nothing "magic" about it - Centera has been employing the same type of object-level replication for years. Only EMC's engineers have figured out how to do RAID protection instead of mirroring to keep the hardware costs low while not sacrificing availability.
I agree that IBM XIV was not the first to do an object-level architecture, but it was one of the first to apply object-level technologies to the particular "use case" and "intended workload" of Web 2.0 applications.
RAID-5 based EMC Centera was designed insteadto hold fixed-content data that needed to be protected for a specific period of time, such as to meet government regulatory compliance requirements. This is data that you most likelywill never look at again unless you are hit with a lawsuit or investigation. For this reason, it is important to get it on the cheapest storage configuration as possible. Before EMC Centera, customers stored this data on WORM tape and optical media, so EMC came up with a disk-only alternative offering.IBM System Storage DR550 offers disk-level access for themost recent archives, with the ability to migrate to much less expensive tape for the long term retention. The end result is that storing on a blended disk-plus-tape solution can help reduce the cost by a factor of 5x to 7x, making RAID level discussion meaningless in this environment. For moreon this, see my post [OptimizingData Retention and Archiving].
While both the Centera and DR550 are based on SATA, neither are designed for Web 2.0 platforms.When EMC comes out with their own "me, too" version, they will probably make a similar argument.
IBM XIV Nextra is not a DS8000 replacement
Nextra is anything but Enterprise-class storage, much less a DS8000 replacement. How silly of all those folks to suggest such a thing.
I did searches on the Web and could not find anybody, other than EMC employees, who suggested that IBM XIV Nextra architecture represented a replacement for IBM System Storage DS8000. The IBM XIV press release does not mentionor imply this, and certainly nobody I know at IBM has suggested this.
The DS8000 is designed for a different "use case" andset of "intended workloads" than what the IBM XIV was designed for. The DS8000 is the most popular disk systemfor our IBM System z mainframe platform, for activities like Online Transaction Processing (OLTP) and large databases, supporting ESCON and FICON attachment to high-speed 15K RPM FC drives. Web 2.0 customers that might chooseIBM XIV Nextra for their digital content might run their financial operations or metadata search indexes on DS8000.Different storage for different purposes.
As for the opinion that this is not "enterprise class", there are a variety of definitions that refer to this phrase.Some analysts look at "price band" of units that cost over $300,000 US dollars. Other analysts define this as beingattachable to mainframe servers via ESCON or FICON. Others use the term to refer to five-nines reliability, havingless than 5 minutes downtime per year. In this regard, based on the past two years experience at 40 customer locations,I would argue that it meets this last definition, with non-disruptive upgrades, microcode updates and hot-swappable components.
By comparison, when EMC introduced its object-level Centera architecture, nobody suggested it was the replacement for their Symmetrix or CLARiiON devices. Was it supposed to be?
Given drive growth rates have slowed, improving utilization is mandatory to keep up with 60-70 percent CAGR
Look around you, Tony- all of your competitors are implementing thin provisioning specifically to drive physical utilization upwards towards 60-80%, and that's on top of RAID 5/RAID 6 storage and not RAID 1. Given that disk drive growth rates and $/GB cost savings have slowed significantly, improving utilization is mandatory just to keep up with the 60-70% CAGR of information growth.
Disk drive capacities have slowed for FC disk because much of the attention and investment has been re-directed to ATA technology. Dollar-per-GB price reduction is slowing for disks in general, as researchers are hitting physicallimitations to the amount of bits they can pack per square inch of disk media, and is now around 25 percent per year.The 60-70 percent Compound Annual Growth Rate (CAGR) is real, and can be even growing faster for Web 2.0providers. While hardware costs drop, the big ticket items to watch will be software, services and storage administrator labor costs.
To this end, IBM XIV Nextra offers thin provisioning and differential space-efficient snapshots. It is designed for 60-90 percent utilization, and can be expanded to larger capacities non-disruptively in a very scalable manner.
On his The Storage Architect blog, Chris Evans wrote [Twofor the Price of One]. He asks: why use RAID-1 compared to say a 14+2 RAID-6 configuration which would be much cheaper in terms of the disk cost? Perhpaps without realizing it, answers itwith his post today [XIV part II]:
So, as a drive fails, all drives could be copying to all drives in an attempt to ensure the recreated lost mirrors are well distributed across the subsystem. If this is true, all drives would become busy for read/writes for the rebuild time, rather than rebuild overhead being isolated to just one RAID group.
Let me try to explain. (Note: This is an oversimplification of the actual algorithm in an effortto make it more accessible to most readers, based on written materials I have been provided as partof the acquisition.)
In a typical RAID environment, say 7+P RAID-5, you might have to read 7 drives to rebuild one drive, and in the case of a 14+2 RAID-6, reading 15 drives to rebuild one drive. It turns out the performance bottleneck is the one driveto write, and today's systems can rebuild faster Fibre Channel (FC) drives at about 50-55 MB/sec, and slower ATA disk at around 40-42 MB/sec. At these rates, a 750GB SATA rebuild would take at least 5 hours.
In the IBM XIV Nextra architecture, let's say we have 100 drives. We lose drive 13, and we need to re-replicate any at-risk 1MB objects.An object is at-risk if it is the last and only remaining copy on the system. A 750GB that is 90 percent full wouldhave 700,000 or so at-risk object re-replications to manage. These can be sorted by drive. Drive 1 might have about 7000 objects that need re-replication, drive 2might have slightly more, slightly less, and so on, up to drive 100. The re-replication of objects on these other 99 drives goes through three waves.
Select 49 drives as "source volumes", and pair each randomly with a "destination volume". For example, drive 1 mapped todrive 87, drive 2 to drive 59, and so on. Initiate 49 tasks in parallel, each will re-replicate the blocks thatneed to be copied from the source volume to the destination volume.
50 volumes left.Select another 49 drives as "source volumes", and pair each with a "destination volume". For example, drive 87 mapped todrive 15, drive 59 to drive 42, and so on. Initiate 49 tasks in parallel, each will re-replicate the blocks thatneed to be copied from the source volume to the destination volume.
Only one drive left. We select the last volume as the source volume, pair it off with a random destination volume,and complete the process.
Each wave can take as little as 3-5 minutes. The actual algorithm is more complicated than this, as tasks complete early the source and volumes drives are available for re-assignment to another task, but you get the idea. XIV hasdemonstrated the entire process, identifying all at-risk objects, sorting them by drive location, randomly selectingdrive pairs, and then performing most of these tasks in parallel, can be done in 15-20 minutes. Over 40 customershave been using this architecture over the past 2 years, and by now all have probably experienced at least adrive failure to validate this methodology.
In the unlikely event that a second drive fails during this short time, only one of the 99 task fails. The other 98 tasks continue to helpprotect the data. By comparison, in a RAID-5 rebuild, no data is protected until all the blocks are copied.
As for requiring spare capacity on each drive to handle this case, the best disks in production environments aretypically only 85-90 percent full, leaving plenty of spare capacity to handle re-replication process. On average,Linux, UNIX and Windows systems tend to only fill disks 30 to 50 percent full, so the fear there is not enough sparecapacity should not be an issue.
The difference in cost between RAID-1 and RAID-5 becomes minimal as hardware gets cheaper and cheaper. For every $1 dollar you spend on storage hardware, you spend $5-$8 dollars managing the environment. As hardware gets cheaper still, it might even be worth making three copies of every 1MB object, the parallel processto perform re-replications would be the same. This could be done using policy-based management, some data gets triple-copied, and other data gets only double-copied, based on whether the user selected "premium" or "basic" service.
The beauty of this approach is that it works with 100 drives, 1000 drives, or even a million drives. Parallel processingis how supercomputers are able to perform feats of amazing mathematical computations so quickly, and how Web 2.0services like Google and Yahoo can perform web searches so quickly. Spreading the re-replication process acrossmany drives in parallel, rather than performing them serially onto a single drive, is just one of the many uniquefeatures of this new architecture.
Wrapping up my week's theme on IBM's acquisition XIV, we have gotten hundreds of positive articles and reviews in the press, but has caused quite a stir with the[Not-Invented-Here] folks at EMC.We've heard already from EMC bloggers [Chuck Hollis] and [Mark Twomey].The latest is fellow EMC blogger BarryB's missive [Obligatory "IBM buys XIV" Post], which piles on the "Fear, Uncertainty and Doubt" [FUD], including this excerpt here:
In a block storage device, only the host file system or database engine "knows" what's actually stored in there. So in the Nextra case that Tony has described, if even only 7,500-15,000 of the 750,000 total 1MB blobs stored on a single 750GB drive (that's "only" 1 to 2%) suddenly become inaccessible because the drive that held the backup copy also failed, the impact on a file system could be devastating. That 1MB might be in the middle of a 13MB photograph (rendering the entire photo unusable). Or it might contain dozens of little files, now vanished without a trace. Or worst yet, it could actually contain the file system metadata, which describes the names and locations of all the rest of the files in the file system. Each 1MB lost to a double drive failure could mean the loss of an enormous percentage of the files in a file system.
And in fact, with Nextra, the impact will be across not just one, but more likely several dozens or even hundreds of file systems.
Worse still, the Nextra can't do anything to help recover the lost files.
Nothing could be further from the truth. If any disk drive module failed, the system would know exactly whichone it was, what blobs (binary large objects) were on it, and where the replicated copies of those blobs are located. In the event of a rare double-drive failure, the system would know exactly which unfortunate blobs were lost, and couldidentify them by host LUN and block address numbers, so that appropriate repair actions could be taken from remote mirrored copies or tape file backups.
Second, nobody is suggesting we are going to put a delicateFAT32-like Circa-1980 file system that breaks with the loss of a single block and requires tools like "fsck" to piece back together. Today's modern file systems--including Windows NTFS, Linux ext3, and AIX JFS2--are journaled and have sophisticated algorithms tohandle the loss of individual structure inode blocks. IBM has its own General Parallel File System [GPFS] and corresponding Scale out File Services[SOFS], and thus brings a lotof expertise to the table.Advanced distributed clustered file systems, like [Google File System] and Yahoo's [Hadoop project] take this one step further, recognizing that individual node and drive failures at the Petabyte-scale are inevitable.
In other words, XIV Nextra architecture is designed to eliminate or reduce recovery actions after disk failures, not make them worse. Back in 2003, when IBM introduced the new and innovative SAN Volume Controller (SVC), EMCclaimed this in-band architecture would slow down applications and "brain-damage" their EMC Symmetrix hardware.Reality has proved the opposite, SVC can improve application performance and help reduce wear-and-tear on the manageddevices. Since then, EMC acquired Kashya to offer its own in-band architecture in a product called EMC RecoverPoint, that offers some of the features that SVC offers.
If you thought fear mongering like this was unique to the IT industry, consider that 105years ago, [Edison electrocuted an elephant]. To understand this horrific event, you have to understand what was going on at the time.Thomas Edison, inventor of the light bulb, wanted to power the entire city of New York with Direct Current(DC). Nikolas Tesla proposed a different, but more appropriate architecture,called Alternating Current(AC), that had lower losses over distances required for a city as large and spread out as New York. But Thomas Edison was heavily invested in DC technology, and would lose out on royalties if ACwas adopted.In an effort to show that AC was too dangerous to have in homes and businesses, Thomas Edison held a pressconference in front of 1500 witnesses, electrocuting an elephant named Topsy with 6600 volts, and filmed the event so that it could be shown later to other audiences (Edison invented the movie camera also).
Today's nationwide electric grid would not exist without Alternating Current.We enjoy both AC for what it is best used for, and DC for what it is best used for. Both are dangerous at high voltage levels if not handled properly. The same is the case for storage architectures. Traditional high-performance disk arrays, like the IBM System Storage DS8000, will continue to be used for large mainframe applications, online transaction processing and databases. New architectures,like IBM XIV Nextra, will be used for new Web 2.0 applications, where scalability, self-tuning, self-repair,and management simplicity are the key requirements.
(Update: Dear readers, this was meant as a metaphor only, relating the concerns expressed above thatthe use of new innovative technology may result in the loss or corruption of "several dozen or even hundreds of file systems" and thus too dangerous to use, with an analogy on the use of AC electricity was too dangerous to use in homes. To clarify, EMC did not re-enact Thomas Edison's event, no animalswere hurt by EMC, and I was not trying to make political commentary about the current controversy of electrocution as amethod of capital punishment. The opinions of individual bloggers do not necessarily reflect the official positions of EMC, and I am not implying that anyone at EMC enjoys torturing animals of any size, or their positions on capital punishment in general. This is not an attack on any of the above-mentioned EMC bloggers, but rather to point out faulty logic. Children should not put foil gum wrappers in electrical sockets. BarryB and I have apologized to each other over these posts for any feelings hurt, and discussion should focus instead on the technologies and architectures.)
While EMC might try to tell people today that nobody needs unique storage architectures for Web 2.0 applications, digital media and archive data, because their existing products support SATA disk and can be used instead for these workloads, they are probably working hard behind the scenes on their own "me, too" version.And with a bit of irony, Edison's film of the elephant is available on YouTube, one of the many Web 2.0 websites we are talking about. (Out of a sense of decency, I decided not to link to it here, so don't ask)
Yesterday's announcement that IBM had acquired XIV to offer storage for Web 2.0 applicationsprompted a lot of discussion in both the media and the blogosphere. Several indicated thatit was about time that one of the major vendors stepped forward to provide this, and it madesense that IBM, the leader in storage hardware marketshare, would be the first. Others were perhaps confused on what is unique with Web 2.0 applications. What has changed?
I'll use this graphic to help explain how we have transitioned through three eras of storage.
The first era: Server-centric
In the 1950s, IBM introduced both tape and disk systems into a very server-centric environment.Dumb terminals and dumb storage devices were managed entirely by the brains inside the server.These machines were designed for Online Transaction Processing (OLTP), everywhere from bookingflights on airlines to handling financial transfers.
The second era: Network-centric
In the 1980s and 1990s, dumb terminals were replaced with smarter workstations and personalcomputers; and dumb storage were replaced with smarter storage controllers. Local Area Networks (LANs)and Storage Area Networks (SANs) allowed more cooperative processing between users, servers andstorage. However, servers maintained their role as gatekeepers. Users had to go through aspecific server or server cluster to access the storage they had access to. These servers continuedtheir role in OLTP, but also manage informational databases, file sharing and web serving.
The third era: Information-centric
Today, we are entering a third era. Servers are no longer the gatekeepers. Smart workstationsand personal computers are now supplemented with even more intelligent handheld devices, Blackberryand iPhones, for example. Storage is more intelligent too, with some being able to offer file sharingand web serving directly, without the need of an intervening server. The roles of servers have changed,from gatekeepers, to ones that focuses on crunching the numbers, and making information presentableand useful.
Here is where Web 2.0 applications, digital media and archives fits in. These are focused on unstructured data that don't require relational database management systems. So long as the useris authorized, subscribed and/or has made the appropriate payment, she can access the information. With the appropriate schemes in place, information can now be mashed-up in a variety of ways, combined with other information that can render insights and help drive new innovations.
Of course, we will still have databases and online transaction processing to book our flights andtransfer our funds, but this new era brings in new requirements for information storage, and newarchitectures that help optimize this new approach.
Well, it's 2008, which could mark the end to RAID5 and mark the beginnings of a new disk storagearchitecture. IBM starts the year with exciting news, acquiring new disk technology from a smallstart-up called XIV, led by former-EMCer Moshe Yanai. Moshe was ousted publicly in 2001 from hisposition as EMC's VP of engineering, and formed his own company. It didn't take long for EMC bloggersto poke fun at this already. Mark Twomey, in his StorageZilla blog, had mentioned XIV before back in August,[XIV], and again todayin [IBM Buys XIV].
To address the new requirements associated with next generation digital content, IBM chose XIV and its NEXTRA™ architecture for its ability to scale dynamically, heal itself in the event of failure, and self-tune for optimum performance, all while eliminating the significant management burden typically associated with rapid growth environments. The architecture also is designed to automatically optimize resource utilization of all the components within the system, which can allow for easier management and configuration and improved performance and data availability.
"We are pleased to become a significant part of the IBM family, allowing for our unique storage architecture, our engineers and our storage industry experience to be part of IBM's overall storage business," said Moshe Yanai, chairman, XIV. "We believe the level of technological innovation achieved by our development team is unparalleled in the storage industry. Combining our storage architectural advancements with IBM's world-wide research, sales, service, manufacturing, and distribution capabilities will provide us with the ability to have these technologies tackle the emerging Web 2.0 technology needs and reach every corner of the world."
The NEXTRA architecture has been in production for more than two years, with more than four petabytes of capacity being used by customers today.
Current disk arrays were designed for online transaction processing (OLTP) databases. The focus was onusing fastest most expensive 10K and 15K RPM Fibre Channel drives, with clever caching algorithmsfor quick small updates of large relational databases. However, the world is changing, and peoplenow are looking for storage designed for digital media, archives, and other Web 2.0 applications.
One problem that NEXTRA architecture addresses is RAID rebuild. In a standard RAID5 6+P+S configuration of 146GB 10K RPM drives, the loss of one disk drive module (DDM) was recovered by reconstructing the data from parity of the other drives onto the spare drive. The process took46 minutes or longer, depending on how busy the system was doing other things. During this time,if a second drive in the same rank fails, all 876GB of data are lost. Double-drive failures are rare,but unpleasant when they happen, and hopefully you have a backup on tape to recover the data from.Moving to slower, less expensive SATA drives made this situation worse. The drives have highercapacity, but run at slower speeds. When a SATA drive fails in a RAID5 array, it could take severalhours to rebuild, and that is more time exposure for a second drive failure. A rebuild for a 750GBSATA drive would take five hours or more,with 4.5 TB of data at risk during the process if a second drive failure occurs.
The Nextra architecture doesn't use traditional RAID ranks or spare DDMs. Instead, data is carved up into 1MBobjects, and each object is stored on two physically-separate drives. In the event of a DDM loss, allthe data is readable from the second copies that are spread across hundreds of drives. New copies aremade on the empty disk space of the remaining system. This process can be done for a lost 750GB drive in under20 minutes. A double-drive failure would only lose those few objects that were on both drives, so perhaps1 to 2 percent of the total data stored on that logical volume.
Losing 1 to 2 percent of data might be devastating to a large relational database, as this could impactthe entire access to the internal structure. However, this box was designed for unstructuredcontent, like medical images, music, videos, Web pages, and other discrete files. In the event of a double-drivefailure, individual files would be recovered, such as with IBM Tivoli Storage Manager backup software.
IBM will continue to offer high-speed disk arrays like the IBM System Storage DS8000 and DS4800 for OLTP applications, and offer NEXTRA for this new surge in digital content of unstructured data. Recognizing this trend, diskdrive module manufacturers will phase out 10K RPM drives, and focus on 15K RPM for OLTP, and low-speedSATA for everything else.
Update: This blog post was focused on the version of XIV box available as of January 2008 that was built by XIV prior to the IBM acquisition. IBM has since made a major revision, made available August 2008 thataddresses a variety of workloads, including database, OLTP, email, as well as digital content and unstructuredfiles. Contact your IBM or IBM Business Partner for the latest details!
Bottom line, IBM continues to celebrate the new year, while the EMC folks in Hopkington, MA will continue to nurse their hangovers. Now that's a good way to start the new year!
Well, it's the last day of the year, and I will be celebrating the new year soon.In the mean time, I leave you with an interesting triple combo related to information.
Nick Carr in his post [Cleaning the Slate] offers a list of articles he did not have time for in 2007.Of these, I enjoyed the 7-page keynote address[Information, Knowledge, Authority and Democracy] by Hunter R. Rawlings III.He talks about the importance of recorded knowledge, including discussions by the US founding fathers Thomas Jefferson and James Madison, and how information is an essential part of democracy.Here's a brief excerpt:
Following the burning of the Capitol in 1815,President James Madison restored the Library of Congress by purchasing ThomasJefferson’s library for the nation. It was Jefferson’s unique classification scheme that thefirst full-time Librarian of Congress, appointed by Madison, used in reorganizing theLibrary. The United States, embodied in the Congress, was to have the best library inthe world because knowledge was necessary to its fundamental purpose, the creationand protection of liberty.
James Madison believed, in other words, that he lived in a “knowledge age.” In ourmyopic way, we like to think that we invented the knowledge age sometime late in the20th century. We did not. Madison and his contemporaries had complete faith andconfidence in the necessity of what they called “useful knowledge,” which, of course,privileged many things we no longer consider useful, such as the ability to read Latinand Greek and to understand the lessons of ancient history.
...by employing collaborative filtering, you use other people’s time to weed out the things that would waste yours. In fact, Del.icio.us and Stumble Upon polls your friends and people with similar interests for the most crucial sources of information and anything else you might have accident skipped over. If The Wisdom of Crowds has taught us anything, it is that a large group of people is drastically more efficient than you’ll ever be on your own.
Unless you enjoy grinding yourself to the bone, use this principle—whether you call it “crowdsourcing” or otherwise—to stop drinking from the information fire hose. It’s not more information, it’s better information, that distinguishes the real winners in business and life.
Finally, Galacticast presents [A Copyright Carol],a humorous 5-minute parody video on what might happen in the future as a result of lawslike the Canadian Digital Millennium Copyright Act[DMCA].
Yesterday, I was able to get the "Build 650" up and running under Qemu emulation onmy Thinkpad laptop computer. Today, I was able to get my Thinkpad and my XO laptoptalking to each other for a "chat".
The built-in "Chat" activity is one of the many kid-friendly activities included onthe XO laptop for the One Laptop Per Child [OLPC] project.It is also possible for two or more people to share other activities, like editing a textdocument, or browsing the internet.
As they say, emulation is only 95% complete, and this is true in this case as well. My Thinkpaddoes not have a built-in video camera, and for some reason the Qemu emulation does not let mehear any sound, despite specifying "-soundhw es1370" parameter. And lastly, it doesn't have the"mesh network" built-in Wi-Fi capability, just standard 54Mbps 802.1g through my Linksys router.
So, I set both XO and Thinkpad to use the new "xochat.org" jabber server so that the two couldsee each other:
$ sugar_control_panel -s jabber xochat.org
I set my XO nickname to be "TonyP" and my Thinkpad to be "Pearson", and chose blue-orange forthe first, and orange-blue for the second.
The process of starting a chat is similar to other IM systems like IBM Lotus Sametime. You havea neighborhood view that shows all people online using the same jabber server. In my case therewere about 30 or so icons on the screen. From the colors on my XO, I was able to locate my Thinkpad,and invite him to a chat. You can share the chat with everyone on the network, or keep it privatebetween two people. I tried both ways to see the difference.
In a private two-way chat, the first person starts up their Chat activity, and sends an inviteto join to another person. The second person sees a flashing chat bubble on the bottom of thescreen to the left of all the other action bar icons. The difference is that the chat bubble isblue-orange matching the sender, rather than black-and-white of the rest of the icons.
If the recipient happens to be busy doing something else full-screen, like browsing the web, theredoesn't seem to be any interruption. It is only when he goes to "home view" will he see the coloredchat bubble and decide to join or not.
The chat itself colorizes the text to match to color of the participant's icons. Blue for one, and orangefor the other. It two people had identical color schemes I guess it might be hard to tell. Thetext is white, so it is best to choose darker colors for contrast.
A nice feature is that you can save your chat session with the "keep" button on the upper rightpart of the screen, and your dialogue discussion will show up as an entry in the "journal".
Using this technique, it is possible for someone who has one "XO" laptop and one regular computer,or two regular computers, to develop and test applications that involve the sharing aspect of educational opportunities. Chats can be between students, student-to-teacher, or event student-to-mentor.
Continuing my week's theme on the XO laptop from the One Laptop Per Child [OLPC] foundation, I successfully managedto emulate my XO on another system.
Part of what is attractive of the XO laptop is the hardware, the high-resolution200dpi screen, the clever screen that rotates and folds flat into an eBook reader,and the water-tight, dust-proof keyboard. The other part is the software, howthey managed to pack an entire operating system, with useful applications, intoa 1GB NAND flash drive.
The drawback for developers like me is the risk of changing something that breaks the system. For example, my first attempt to create my own activityresulted in a blank space in my action bar, and my journal went into someinfinite loop, blinking as if it were still loading for minutes on end. I fixed it by deleting out the activity I created and rebooting.
To get around this, I successfully ran the disk-image under Linux's Virtual Machinesoftware called Qemu. This is an open source offering, with a proprietary add-onaccelerator called Kqemu. Here were the steps involved:
Base Operating System
Qemu is now available to run on Linux, Windows and OS X-Intel. I have an Ubuntu 7.04"Feisty Faun" version of Linux installed on my system from a project I did last year, so decided to use that.
Normally, "apt-get install qemu" would be enough, but I wanted to get the latest release, so I downloaded the [0.9.0 version]tarball of compiled binaries. Note that trying to compile Qemu from source requiresa downlevel gcc-3.x compiler, and my attempts to do this failed. The compiled binariesworked fine.
The Kqemu author hasn't packaged this for distribution, so I download the source code anddid my own compiles. You can do the "configure-make-install" using the regular gcc 4.1compiler and it went smoothly.
Getting Kqemu active was bit of a challenge. I had to make sense of Nando Florestan's[Installing Kqumu in Ubuntu] article,and the subsequent comments that followed.
There is a tiny [8MB Linux image]that should be used to verify the Kqemu is activated correctly.
The Disk Image
As with other development efforts, there are the older stable versions, and the bleedingedge development versions. I chose the 650 Build from the [Ship.2 stable versions], whichmatches the version on my XO laptop. The image comes as a *.bz2, which is a highly-compressedfile. Using "Bunzip2", the 221MB file expands to something like 932MB.
I renamed the resulting file to "build650.img"
Once I got all this done, I then made a simple script "launch" in my /home/tpearson/bin directory:
Then "launch build650.img" was all I needed to run the emulation. The full-screen mode helpsemulate the view on XO laptop. I was able to change the jabber server to "xochat.org" and see otherXO laptops online on my neighborhood view.
When running under Qemu, you can't just press Ctrl-Alt-something. For example, Ctrl-Alt-Erase onthe XO reboots the Sugar interface. However, do this on a Linux system, and it reboots your nativeX interface, blowing away everything.Instead, you press Ctrl-Alt-2 to get to the Qemu console, designated by (qemu) prompt,and then type:
Press "Ctrl-Alt-1" followed by "Ctrl-Alt" to get back to the emulated XO screen.
With this emulation, I am more likely to try new things, change files around, edit system files,and so on, without worrying about rendering my actual XO laptop unusable. Once debugged, I canthen work on moving them over to my XO, one at a time.
Wrapping up this week's theme on the XO laptop, I decided to take on thechallenge of printing. I managed to print from my XO laptop to my laserjet printer.I checked the One Laptop Per Child [OLPC] website,and found there is no built-in support for printers, but there have been several peopleasking how to print from the XO, so here are the steps I did to make it happen.
(Note: I did all of these steps successfully on my Qemu-emulated system first, and then performed them on my XO laptop)
Step 1: Determine if you have an acceptable printer
The XO laptop can only connect to a printer via USB cable or over the network.Check your printer to see if it supports either of these two options. In my case, my printer is connected to my Linksys hub that offers Wi-Fi in my home.
The XO runs a modified version of Red Hat's Fedora 7, so we need to also determineif the printer is supported on Linux.Check the [Open Printing Database]for the level of support. This database has come up with the following ranking system.Printers are categorized according to how well they work under Linux and Unix. The ratings do not pertain to whether or not the printer will be auto-recognized or auto-configured, but merely to the highest level of functionality achieved.
Perfectly - everything the printer can do is working also under Linux
Mostly - work almost perfectly - funny enhanced resolution modes may be missing, or the color is a bit off, but nothing that would make the printouts not useful
Partially - mostly don't work; you may be able to print only in black and white on a color printer, or the printouts look horrible
Paperweight - These printers don't work at all. They may work in the future, but don't count on it
If your printer only supports a parallel cable connection, or does not have a high enough ranking above, go buy another printer. The [Linux Foundation] websiteoffers a list of suggested printers and tutorials.
In my case, I have a Brother HL5250-DN black-and-white laserjet printer connected over a network to Windows XP, OS X and my other Linux systems. It is rated as supporting Linux perfectly, so I decided to use this for my XO laptop.
Step 2: Install Common UNIX Printing System (CUPS)
Technically, Linux is not UNIX, but for our purposes, close enough. Start the Terminalactivity, use "su" to change to root, and then use "yum" to install CUPS. Yum will automatically determine what other packages are needed, in this case paps and tmpwatch. Once installed, use "/usr/sbin/cupsd" to get the CUPS daemon started, and add this to the end ofrc.local so that it gets started every time you reboot.
Click graphic on the left to see larger view
[olpc@xo-10-CC-6F ~]$ subash-3.2# yum install cups...Total download size = 3.0 MIs this OK [y/N]? y
To download the appropriate drivers, you may need a browser that can handle file downloads. I have triedto do this with the built-in Browse activity (aka Gecko) but encountered problems. I have both Opera and Firefox installed, but I will focus on Opera for this effort.I also installed the older18.104.22.168 version of the Flash player (worked better than the latest 22.214.171.124 version) and Java JRE.Follow the OLPC Wiki instructions for [Opera, Adobe Flash,and Sun Java] installation, thenverify with the following [Java and Flash] testers.
Step 4: Download drivers and packages unique for your printer
In my case, I used Opera to get to the [Brother Linux Driver Homepage], and downloaded the RPM's for LPR and CUPS wrapper. These are the ones listed under "Drivers for Red Hat, Mandrake (Mandriva), SuSE". I saved these under "/home/olpc" directory.
By default, the root user has no password. However, you will need it to be something for later steps,so here is the process to create a root password. I set mine to "tony" which normallywould be considered too simple a password, but ignore those messages and continue.We will remove it in step 8 (below) to put things back to normal.
[olpc@xo-10-CC-6F ~]$ subash-3.2# passwdChanging password for user root.New UNIX password: tonyBAD PASSWORD: it is too shortRetype new UNIX password: tonypasswd: all authentication tokens updated successfullybash-3.2# exit[olpc@xo-10-CC-6F ~]$
Step 6: Launch CUPS administration
Here I followed the instructions in Robert Spotswood's [Printing In Linux with CUPS] tutorial.Launch the Opera browser, and enter "http://localhost:631/admin" as the URL. The localhostrefers to the laptop itself, and 631 is the special port that CUPS listens to from browsers. You can alsouse 127.0.0.1 as a shortcut for "localhost", and can be used interchangeably.
In my case, it detected both of my networked printers, so I selected the HL5250DN, entered thelocation of my PPD file "/usr/share/cups/model/HL5250DN.ppd" that was created in Step 4. I set the URI to "lpd://192.168.0.75/binary_p1" per the instructions [Network Setting in CUPS based Linux system] in the Brother FAQ page. I chage the page size from "A4" to "Letter".I set this printer as the default printer. When it asks for userid and password, that is whereyou would enter "root" for the user, and "tony" or whatever you decided to set your root password to.
Select "Print a Test Page" to verify that everything is working.
Step 7: Printing actual files
Sadly, I don't know Opera well enough to know how to print from there. So, I went over to my trustedFirefox browser. Select File->Page Setup to specify the settings, File->Print Preview tosee what it will look like, and then File->Print to send it to the printer.
To print the file "out.txt" that is in your /home/olpc directory, for example, enter"file:///home/olpc/out.txt" as the URL of the firefox browser. This will show the file,which you can then print to your printer. I had to specify 200% scaling otherwise the fontswere too small to read.
Step 8: Remove the "root" password
If you want to remove the root password, here are the steps.
[olpc@xo-10-CC-6F ~]$ suPassword: tonybash-3.2# passwd -d rootRemoving password for user root.passwd: Successbash-3.2# exit[olpc@xo-10-CC-6F ~]$
Now the problem is that there is no way to print stuff from any of the Sugar activities. The best place toput in print support would be the Journal activity. Along the bottom where the mounted USB keys arelocated could be an icon for a printer, and dragging a file down to the printer ojbect could cause it tobe send to the printer.
The alternative is to write some scripts invocable from the Terminal activity to determine what isin the journal, and send them to LPR with the appropriate parameters.
I did not have time to do either of these, but perhaps someone out there can take on that as a project.
Continuing my week's theme on the XO laptop from the One Laptop Per Child [OLPC] project, I have been amused watching the OLPC forum discussion on the choiceof browser options available.
The built-in browser is simple but functional. It is full screen,with back, forward, and bookmark buttons, and an entry field forthe URL. This browser is fully integrated with the Sugar platform,files downloaded will appear in the journal. Download an Activity*.xo file, for example, and you can install it from the Journal.If you want to upload a file, click BROWSE on the website, and theJournal will pop up to choose files from.
Out of the box, the XO supports a minimal Flash that can handlesome Flash-based games but not YouTube videos, and does not supportJava.
The good folks of Opera have built a special edition for the XO laptop.However, some settings need to be changed to make the fonts large enoughto read.
Opera can be run as a Sugar activity, but this just launches a mothertask, which in turn launches a daughter task that actually runs thebrowser. This means that Home View will have two icons. The mothertask has an the Opera icon, but click on it and you get a grey screen.The daughter task appears as a grey circle, click on it and you get thebrowser screen. Alt-Tab will rotate through the Activities, so thegrey screen of the mother task is part of the rotation.
Although Opera has one foot on the Sugar platform, and one foot off,the lack of integration means poor interaction with the journal. The use of Opera is correctly registered. However, downloadingfiles requires a working knowledge of subdirectories, and uploading anythingrequires knowing what it is called, and where it is located. Not obviousfor many of the items created by Sugar applications.
The XO laptop is based on Redhat Fedora distribution, so I downloadedthe Firefox RPM package and installed this. To run, you need to startthe Terminal Activity, and then at the cursor type firefox.Journal only registers that the Terminal activity was used, but not anythingelse.
Since I run Firefox 2.0 on Windows XP, OS X and Linux, I am very familiarwith this browser, and it works as expected. Like Opera, there are shortcut keys, tabs for multiple pages, and optionsto add Java and Flash player. I was able to install add-onsfor Del.icio.us and FireFTP, and they worked as expected. Having accessto FTP sites will make development on the XO much easier.Again all files are uploaded/downloaded to directories, so some workingknowledge of where files are placed is required.
The fonts in Firefox did not expand/shrink as nicely as they had in Opera.Be careful not to select "View->
To close, you have to select File->Quit from the browser window, whichbrings you back to the Terminal activity, which you can then shutdown with Ctrl-Esc.
For now, I will keep all three and continue to evaluate them.I saw a few opportunities for improvement:
The Opera and Terminal icons are not on the first screen.You have to hit the right arrow to get to the "overflow" set of icons. Re-ordering the icons is simply a matter of editing the following file with "vi"(my first few lines I use are shown below):
Put the activities in the order you want. Any activity not listed willappear after these.
It might be possible to create a modified Terminal activity thatinvoked Firefox directly, to eliminate having to type it in each time.
Several people have expressed interest in a browser that runs entirely withthe Xo laptop folded over in eBook/Game mode, such that thekeyboard is completely covered up, exposing only the up-left-right-down arrowsand the Circle/Square/X/Check buttons.
Change the "News Reader" to invoke Bloglines instead. This might be yetanother modified Terminal activity, but borrow the icon from News.
Well, if you have further thoughts on these browsers, enter a comment below.
My XO laptop arrived Friday, December 21, this was from the [Give 1 Get 1 (G1G1)] program fromthe One Laptop Per Child (OLPC) foundation. The program continuesto the end of this month (December 31).
Here are my first impressions.
Setup was Easy
Open the box, put in battery, and plug in the adapter. Enter your name and choose your favorite color for your stick figurine. No passwords, no parameters. Software is pre-installed and ready to use.
The four pages of instructions included how to open the unit (not intuitive), where the various connection ports are located, what the home screen and neighborhood screen look like, safety warnings, and a nice letter from Nicholas Negroponte with an 800 phone number and website in case more help is needed.
Connecting to the internet was the first thing I did. The neighborhood screen shows all the Wi-Fi access points. It recognized mineand three others. I clicked on mine, entered my WEP key, and was connected.
This is a Linux operating system running the Sugar user interface.There are four screens:
Neighborhood - shows all Wi-Fi access points
Friends - shows all other XO laptops nearby, in my case I am all alone
Home - your stick figurine with all the applications you can choose from are represented as icons at the bottom, just like OS X on my Mac Mini, or the launchpad on my Windows XP. Left panel for clipboard items.
Application - Applications run in full-screen mode
Four buttons across the top allow you to jump to any screen instantly.Everything else is single left-click. No double-clicks or right-clicks.
A circle on the home screen designates which applications are running, and how much of the available 256MB RAM they are consuming. This makes it easy to seeif you can run more applications or need to shut something down. Youcan jump to any application, or shut it down, from this view.
Shutting down the XO is done by clicking your stick figurine,and choosing shutdown.
I fired up the browser. The default 'home page' offers some help offline, as well as links to online resources and a google search bar. The full-color 1200x900 is very easy to read. You can hit ctrl+plus to make the fonts bigger. In bright sunlight, the screen turns automatically to greyscale.The built-in browser is easy enough to use, with standard back, forward, re-load, and bookmark buttons. The URL entry field also shows the pages title. It doesn't have tabs to see multiple pages at the same time, but I was able to fire up a second instance of the browser, so thatI could alt-tab back and forth between the two web sites.
There are so many applications that they don't all fit on the bottom of the screen.Left and right tab buttons will display the next set. I don't know if it is possible to re-order the icons, but I can certainly see some applications appealing to different ages, and perhaps re-ordering them into age-specific groups might be helpful.
Basic applications include the Abiword word processor, a PDF viewer, a simple paint program, calculator, chat, and news RSS feed reader; TamTam music to play and edit compositions; and some learn-to-program-a-computer software including Pippy, Etoys, and TurtleArt.
The 'record' program lets you take 640x480 pictures with the built-in camera, up to 45 seconds of video and audio recording. The picture abovewas taken with my XO, and edited online using [snipshot.com]. Another program can be usedto make video calls to another computer, similar to Skype or IBM Lotus Sametime.
The XO has built-in microphone and speakers, but also microphone and speaker ports, as well as three USB ports, and a slot for an SD memory card.
The QWERTY keyboard is designed for small children hands, I found myself using my two index fingers in a hunt-and-peck style. People who use Blackberry's or other hand-held devices might be able to use their two thumbs instead. Also, I am not used to a touchpad as the pointing device. My other laptops have a red knob between the G/H/B keys that acts like a joystick. So, I decided to attach my Apple keyboard/mouse to one USB port, which allows me faster typing and better resolution with my mouse.
I also inserted a 1GB SD card into the slot. Getting to the SD slot was challenging--you have to rotate the screen 90 degrees so that the lower right corner is over the laptop handle. It appears I need to purchase some tweasers to get my SD card back out, so until then, it will remain there as permanent addition to my XO.
A terminal application provides a command line interface into Linux.
The 'vi' editor is installed, in case I need to make changes to fstab or anythingelse in my /etc directory.
There is no S-video or VGA port. However, a teacher could probably fold thislaptop up in e-book mode and lay it flat on an [overhead projector] since the screen can handle bright sunlight in black-and-white mode.
The Journal and the Clipboard
There are no folders or subdirectories here. The journal acts as your desktop, holding all the files you have referenced, sorted in chronological order with the most recent on top. The journal application is started automatically when you boot up.My SD card is shown as a separate entry at the bottom right corner, but I have access only to files on my top-level directory on the card. The journal allows you to drag and drop between the system and the SD flash card.The list can be filtered by file type and application, so finding things is easy.You can also copy anything in the journal to the clipboard, appearing on the leftpanel of the home screen. You can then launch or paste this into other applications.
Pressing Alt-1 takes a 1200x900 snapshot of the current screen, and puts it into the journal.On websites that allow you to upload a file, including GMAIL, snipshot.com, etc. the browse button brings up the journal. So, for example, you could take a snapshot of the current webpage or paint creation, and send it as an attachment to someone via GMAIL. Google has an XO-enabled version of GMAIL that you can download from the OLPC activities page.
This entire post, including the picture above, was done with the XO laptop itself. I am impressed with the thought that went into this design, and I see great potential here. The interface adequately hides the Linux operating system for those who just want to use the computer, but makes it readily accessible for those who want to learn more about the Linux operating system and computer programming.
Well, tomorrow is the Winter solstice, at least for those of us in the Northern hemisphere of the planet.As often happens, I have more vacation days left than I can physically take before they evaporateat the end of the year, so next week I will be off, going to see movies like the new["Golden Compass"]or perhaps read the latest book from [Richard Dawkins].
Next week, I suspect some of the kids on my block will be playing with radio-controlled cars orplanes. If you are not familiar with these, here's a [video on BoingBoing]that shows Carl Rankin's flying machines that he made out of household materials.
Which brings me to the thought of scalability. For the most part, the physics involvedwith cars, planes, trains or sailboats apply at the toy-size level as well as the real-world level. One human operator can drive/manage/sail one vehicle. While I have seen a chess master play seven opponents on seven chess boards concurrently, itwould be difficult for a single person to fly seven radio-controlled airplanes at the same time.
How can this concept be extended to IT administrators in the data center? They have to deal withhundreds of applications running on thousands of distributed servers.In a whitepaper titled [Single System Image (SSI)], the threeauthors write:
A single system image (SSI) is the property of a systemthat hides the heterogeneous and distributed nature of theavailable resources and presents them to users and applicationsas a single unified computing resource.
IBM has some offerings that can help towards this goal.
Even in the case where yourvehicle is being pulled by eight horses--(or eight reindeer?)--a single operator can manage it, holding the reins in both hands. In the same manner,IBM has spent a lot of investment and research into supercomputers, where hundreds of individualservers all work together towards a common task. The operator submits a math problem, for example,and the "system system image" takes care of the rest, dividing the work up into smaller chunksthat are executed on each machine.
When done with IBM mainframes, it is called a Parallel Sysplex. The world's largest business workloadsare processed by mainframes, and connecting several together and working in concert makes this possible.In this case, the tasks are typically just single transactions, no need to divide them up further, justbalance the workload across the various machines, with shared access to a common database and storageinfrastructure so they can all do the work equally.
Last August, in my post [Fundamental Changes for Green Data Centers], I mentioned that IBM consolidated 3900 Intel-based servers onto 33 mainframes. This not only saves lots of electricity, but makes it much easier for the IT administratorsto manage the environment.
Parallel Sysplex configurations often require thousands of disk volumes, which would have been quitea headache dealing with them individually. With DFSMS, IBM was able to create "storage groups" wherea few groups held the data. You might have reasons to separate some data from others, put them inseparate groups. An IT administrator could handle a handful of storage groups much easier than thousandsof disk volumes. As businesses grow, there would be more data in each storage group, but the numberof storage groups remains flat, so an IT administrator could manage the growth easily.
IBM System Storage SAN Volume Controller (SVC) is able to accomplish this for other distributed systems.All of the physical disk space assigned to an SVC cluster is placed into a handful of "managed diskgroups". As the system grows in capacity, more space is added to each managed disk group, but few IT administrators can continue to manage this easily.
The new IBM System Storage Virtual File Manager (VFM) is able to aggregate file systems into one globalname space, again simplifying heterogeneous resources into a single system image. End users have a singledrive letter or mount point to deal with, rather than many to connect to all the disparate systems.
Lastly we get to the actual management aspect of it all. Wouldn't it be nice if your entire data centercould be managed by a hand-held device with two joysticks and a couple of buttons? We're not quite there yet, but last October we announced the [IBM System Storage Productivity Center (SSPC)]. This is a master consolethat has a variety of software pre-installed to manage your IBM and non-IBM storage hardware, includingSAN fabric gear, disk arrays and even tape libraries. It lets the storage admin see the entire data centeras a single system image, displaying the topology in graphical view that can be drilled down using semanticzooming to look at or manage a particular device or component.
Customers are growing their storage capacity on average 60 percent per year. They could do this by havingmore and more things to deal with, and gripe about the complexity, or they can try to grow theirsingle system image bigger, with interfaces and technologies that allow the existing IT staff to manage.
As we wrap up the year, people's thoughts turn to archive anddata retention.
The [Robert Frances Group] have put out a research paper titled Optimizing Data Retention and Archiving - November 2007 that helps IT executives understand the cost differences for a disk-only archive approach versus disk/tape archive approach and how an [IBM System Storage DR550] offering can help address the long-term storage archive requirements with a world-class storage strategy that reduces cost, improves efficiency and supports compliance. Here is an excerpt:
Ongoing legal, audit, and regulatory requirementswill continue to drive IT groups to improvearchive policies, processes, strategy, andefficiency. The choice of which technologies touse will have a profound impact on the success ofsuch efforts, since technologies like the DR550embody many aspects of the strategy, processes,and policies that must be decided upon. When itcomes to tape, IBM's DR550 is unique inproviding that support. Competitors tout disk-onlysolutions as the wave of the future, but researchindicates otherwise. The most basic benefits arecost and mobility, and despite the various vendorproclamations to the contrary, tape is still only afraction of the cost of disk and will remain so inthe foreseeable future.
This paper is yet another nail in the coffin of EMC Centera.In his post [Anyone Naughty on Your List…], Jon W Toigo points to an eBay fire sale of an EMC Centera Gen 4.
There has never been a better time to switch from EMC Centera to theIBM System Storage DR550.
This is a reasonable question. Since Invista 2.0 came out months ago in August, and Invista 2.1 is rumored to be out by end of this month, why put out a press release now, rather than just wait a few weeks? Thesignificant part of this announcement was that EMC finally has their first customer reference.To be fair, getting a customer to agree to be a reference is difficult for any vendor. Some non-profitsand government agencies have rules against it, and some corporations just don't want to be bothered byjournalists, or take phone calls from other prospective customers. I suspect EMC wanted to put the good folks from Purdue University in front of the cameras and microphones before they:
In Moore's terminology, Purdue University would be a "technology enthusiast", interested in exploring the technologyof the EMC Invista. Universities by their very nature often see themselves as early adopters, willing to take big risks in hopes to reap big rewards. The chasm happens later, when there are a lot of early adopters, all willing to be reference accounts. The mainstream market--shown here as pragmatists, conservatives, and skeptics-- are unwillingto accept reference claims from early adopters, searching instead for moderate gains from minimal risks. They prefer references from customers that are similar in size and industry. Whether a vendor can get a product to cross this chasm is the focus of the book.
Why "SAN" virtualization?
Technically, Invista is "storage" virtualization, not "SAN" virtualization. Virtualizationis any technology that makes one set of resources look and feel like a different setof resources, preferably with more desirable characteristics. You can virtualizeservers, SANs, and storage resources.
Virtual SAN (VSAN) technology, supported bythe Cisco MDS 9500 Series Multilayer Director Switch, partitions a single physical SAN into multipleVSANs, allowing different business functions and requirements to share a common physical infrastructure.
How does Invista advance Cisco's VSAN functionality? It doesn't, but that doesn't makethe title a falsehood, or the press release by association full of lies.If you read the entire press release, EMCcorrectly states that Invista is "storage" virtualization. Some storagevirtualization products, like EMC Invista and IBM System Storage SAN Volume Controller (SVC), require a SAN as a platform for which to perform their magic.Marketing people might use the term "SAN" torefer not just the network gear that provides the plumbing, but also to include the storage devices that are attached to the SAN. In that light, theuse of "SAN virtualization" can be understood in the title.
More importantly, it appears that EMC no longer requires that you purchase new SAN equipment from themwith Invista. When the Invista first came out, it cost over a quarter-million US dollars to cover thecost of the intelligent switches, but with the price drop to $100K, I imagine this means theyassume everyone has an appropriately-supported intelligent switch already deployed.
Why this architecture?
In his post [Storage Virtualization and Invista 2.0], EMC blogger ChuckH does a fair job explaining why EMC went in this direction for Invista, and how it is different thanother storage virtualization products.
Most storage virtualization products are cache-based. The world's first disk storagevirtualization product, the IBM 3850 Mass Storage System, introduced in 1974, and thefirst tape virtualization product, the IBM 3494 Virtual tape Server, introduced in 1997, bothused disk cache in front of tape storage. Later virtualization products, like IBM SVC and HDS USP-V, use DRAM memory cache in front of disk storage, but the concept is the same.People are comfortable with cache-based solutions, because the technology is matureand well proven in the marketplace, and excited and delighted that these can offer the following features in a mixed heterogeneous disk environment:
instantaneous point-in-time copy
None of these features are provided by Invista, as there is no cache in the switch. Instead,Invista is a "packet cracker"; it cracks open each FCP packet, inspects and modifies the contents, then passes theFCP packet along to the appropriate storage device. This process slows down each read andwrite by some amount, perhaps 20 microseconds. The disadvantage of slowing down every readand write is offset by having other benefits, like non-disruptive data migration.
To compensate for Invista's inability to provide these features,EMC offers a second solution called EMC RecoverPoint, which is an in-band cache-based appliancesimilar in design to SVC, but maps all virtual disks one-to-one to physical disks. It offersremote distance asynchronous mirroring between heterogeneous devices.EMC supports RecoverPoint in front of Invista, but if you are considering buying bothto get the combined set of features, you might as well buy an IBM SVC or HDS USP-V instead,in one system, rather than two, which is much less complicated. IBM SVC and HDS USP-Vhave both "crossed the chasm" having sold thousands of units to every type and size of customer.
Hopefully, this answers the questions you might have about EMC Invista.
Last year, I covered Chris Anderson's book [The Long Tail]. This year, Chris Anderson, editor-in-chief of Wired.com, has an upcoming book titled Free, the past and future of a radical price. Chris talked about his book here at Nokia World 2007 conference, and the [46-minute video] is worth watching.He asks the big question "What if certain resources were free?" This could be electricity, bandwidth, or storage capacity. He explores how this changes the world, and createsopportunities for new business models. However, many people are stuck in a "scarcity" modeland treat nearly-free resources as expensive, and find themselves doing traditional things thatdon't work anymore. Chris mentions [Second Life] as aneconomy where many resources are free, and seeing how people respond to that.Rather than focusing on making money, new businesses are focused on gainingattention and building their reputation. Here are some example business models:
Cross-subsidy: give away the razors, sell the razor blades; or give away cell phones and sell minutes
Ad-Supported: magazines and newspapers sell for less than production costs
Freemium: 99% use the free version, but a handful pay extra for something more
Digital economics: give away digital music to promote concert tours
Free-sample marketing: give away samples to get word-of-mouth advertising
Gift economy: give people an opportunity and platform to contribute like Wikipedia
Nick Carr writes a post [Dominating the Cloud], indicatingthat IBM, Google, Microsoft, Yahoo and Amazon are the five computing giants to watch, as they are more efficient atconverting electricity into computing than anyone else. Last month, I mentioned IBM and Google partnership on cloud computing in my post[Innovationthat matters: cell phones and cloud computing].Nick's upcoming book titled[The Big Switch] looks into "Utility Computing",comparing the change of companies generating their own electricity to using an electric grid, to the recent developments of cloud computing and software as a service (SaaS). Amazon's latest "SimpleDB" online databaseis cited as an example.
Last, but not least, Seth Godin writes in his post [Meatballs and Permeability] about the bits-vs-atoms issue, what Chris Anderson above refers to as the new digital economy. The idea here is that value carried electronically as bits (digital documents, for example) have completely different economics than value carried as atoms (physical objects), andrequires new marketing techniques. Methods from traditional marketing will not be effective in this new age.Here is a [review] of Seth's new book Meatball Sundae: Is Your Marketing Out of Sync?
All three of these books seem to be covering the same phenomenon, just from different viewpoints. I lookforward to reading them.
While Bill Gates is personally benefiting from code he wrote 30 years ago, most software engineers don't getroyalties for their creative efforts. Robin Harris on StorageMojo has a great piece on [Why are the writers striking?]The writers in this case are those who write scripts for television programs. They get 4 cents for every$19.99 DVD sold today, and want this bumped up to 8 cents. More importantly, they want the same deal forcontent shown over the internet. Currently, they get nothing when content they wrote for is shown on the internet, and they would like that fixed also.
Paying royalties to creative writers encourages them to write good stuff. The best stuff will result in moreroyalties, and we want to encourage this. What about software engineers? Don't we want them to write the beststuff also? Shouldn't they get royalties too, not just a flat salary and continued employment?
From a technology-oriented to a service-oriented approach to IT management
Companies are challenged with shifting from a technology/resource-oriented to a service-oriented approach to IT management. This involves new processes, a new reportingstructure for the IT staff, new tools and technologies, and new data to be captured.A top-down approach is recommended for large organizations, but a bottom-up approachmight be easier to implement for small and medium sized businesses.
IT service management architecture and autonomic computing
IBM has been promoting the concept of Autonomic Computing since 2001. A self-managed resource can have an autonomic manager with sensor and effector. The sensor is used to monitor status, a knowledge basecan analyze and plan for appropriate modifications, and execute these through theeffector. The Autonomic Computing Reference Architecture (ACRA) aligns with the Information Technology Service Management (ITSM) model well, with the CMDB acting asthe knowledge base for the autonomic managers. See my earlier post[Self tuning guitars and storage].
Evolving standards for IT service management
Changes to the IT infrastructure must be closely managed to avoid disruptions.IT organizations recognize that standards-based solutions enable interoperability,with less risk, to connect internal and external applications. Standards can be formally developed by standards bodies like ISO, IETF, W3C, OASIS, and DMTF; or be de facto standards that become widely used by companies, which can then laterbe adopted by standards bodies. SML and SDD are emerging standards that are incompatible with the current set of Web Services-based protocols, like WSDM, but work isunderway to try to determine a unifying standard to support all of these under ITSM.
Prospects for simplifying ITSM-based management through self-managing resources
An ideal computing system would take over a great deal of its own management.Today's IT systems are brittle, difficult to understand, and dangerous to change.The savings from automating some tasks are dwarfed by the irreducible costs of humandecision making, agreements and approvals built in formal processes. A true self-managing, scalable IT system would consist of a number of nearly-identical boxes,with a web interface to define high-level policies and provide information on utilization and performance. As the system needs to expand, it can automatically place the order. When the new boxes arrive, they are placed and connectedinto the data center, and the system configures and provisions them appropriately.
IT Autopilot: A flexible IT service management and deliver platform for smalland medium business
Using an airplane analogy, the pilot performs manual steps to get the plane safelyoff the ground, then turns it over to the autopilot for normal operations. The ITAutopilot intends to do this for IT service management in small and medium business (SMB)that may not have a large dedicated IT staff, using an SOA approach that isloosely coupled, stateless, and adhering to Web Services standards. The IT Autopilotemploys workflow-based controls, the autonomic computing MAPE model, and customizedpolicies to address SMB requirements. It could be deployed as an appliance, similarto IBM System Storage Productivity Center.
I'm continuing my coverage of IBM Systems Journal's [fifteen articles about IBM Service Management].As storage hardware cost per GB declines 25 percent per year, the cost of labor has grown to nearly 70percent of the total IT budget. This brings new focus on how we do things, rather than what things siton the raised floor. Yesterday, my post summarized[the first five articles].Here is what I got out of the next five articles:
Integrated change and configuration management
IT Infrastructure Library (ITIL) best practice covers a variety of disciplines, including incident management,problem management, release management, service help desk, change management, and configuration management.IBM has combined the last two into a single database, and this paper provides insights gained fromimplementing these in practice. A special section talks about how service providers can support multipleclients that must be kept separate from each other.
The process of building a Process Manager: Architecture and design patterns
Business processes coordinate and sequence the work done by a collection of people.Most companies define their business process from scratch, and develop their own applicationsto support their implementation. Process Managers are "out of the box" applications that help customers integrateand automate more quickly than building from scratch. These Process Managers leverage and update informationabout configuration items (CIs) in the configuration management database (CMDB). One of the first developedby IBM was the IBM Tivoli Storage Process Manager.
Integration of domain-specific IT processes and tools in IBM Service Management
ITIL tells you what needs to get done, but it doesn't tell you exactly how to do it. Completing a simplechange request to the IT environment can have a drastic impact on service level agreements (SLAs), utilization of existing storage capacity, and business operations. Sometimes it is important to use multipleProcess Manager applications together. To accomplish this, it is important to launch and land in contextat the appropriate points for smooth transition.
Using a model-driven transformational approach and service-oriented architecture for service deliver management
Companies are considering outsourcing as a way to focus on core competencies. However, the trend is towardselective outsourcing, where the customer controls the IT solution architecture and retains their legacy tools.As a result, service providers inherit the business and IT processes from their clients. IBM Research has developed the model-driven business transformation (MDBT) method that choreographs workflow tools with humanactivities. A "balanced scorecard" allows both client and outsourcer monitor progress towards strategic goals.
Catalog-based service request management
Service providers (outsourcers) are able to bring the latest IT technology, best practices, and skilledservice delivery teams. Unfortunately, unique business processes from each client limits the ability to leveragethese resources effectively. A service delivery management platform (SDMP) catalog serves as a repositoryof atomic services and the delivery teams that can perform them. This allows outsourcers to leverage resourcesacross multiple clients, while still being able to tailor business compositions of these atomic services to an individual client's requirements.
The latest IBM Systems Journal has [fifteen articles about IBM Service Management], which includes the disciplines for managing storage resources as part of an overall IT data center.As with most journals, these articles are heavy academic efforts, not light summer reading.
However, since I have moved from marketing to consulting, I need to read these kinds of articles to keep up with the industry. I realize many people don't have time to read allof these, so over the next three days, I will give some quick highlights in hopefully more understandablelanguage. Here is what I got out of the first five articles:
An Overview of IBM Service Management
This 10-page article provides a good overview of what the other articles go into greater detail.The role of information has changed, from supporting back-office tasks like payroll andinventory, to enabling growth in the business itself, providing insight and competitive advantage. The challenges are summarized under "Four C's": Complexity, Change, Cost, and Compliance. The recommended approach is to engage with IBM,who has thousands of practitioners with years of experience in ITIL, eTOM, COBIT, CMMI and SOA.
Adding value to the IT organization with the Component Business Model
Many Service Level Agreements (SLAs) are loaded with technological jargon rather than concentratingon intended business results. CIOs must change this, and learn to run IT as a business witha service delivery focus.IBM Process Reference Model for IT (PRM-IT) is the foundationfor the Component Business Model for the Business of IT (CBMBoIT) that can assist with strategic decision making to transform IT into this new role.
An Integration model for organizing IT service management
There are so many ways to implement Information Technology Service Management (ITSM) that it is hard to tell if there are gaps or overlaps between products and offerings. A seamless solution requires common terminology and approaches. An integration model helps to bring all this together, focusedon being consistent with existing practice, with clarity of expression, and practical to implement.
IBM Service Management architecture
Today's systems management tools are fragmented by resource domain--servers are managed here, networksmanaged there, and storage is another story altogether. IBM Service Management intends to integratea portal-based User Interface, a process runtime layer, a configuration management database (CMDB), and all the various operational management products (OMPs) for each resource. For example, IBM TotalStorage ProductivityCenter is an OMP for IBM and non-IBM storage resources.
A configuration management database architecture in support of IBM Service Management
IBM Tivoli Change and Configuration Management Database (CCMDB) holds all the configuration data of IT resources in the data center, including individual "configuration items" (CIs), as well as tracks changes. The database is populated with data from different sources, includingautomatic discovery. Relationships between CIs provides a visual representation of application dependencies.The data model uses a clever combination of Unified Modeling Language (UML) with Java persistent objects.