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 a Master Inventor and Senior IT Specialist for the IBM System Storage product line at the
IBM Executive Briefing Center in Tucson Arizona, and featured contributor
to IBM's developerWorks. In 2011, Tony celebrated his 25th year anniversary with IBM Storage on the same day as the IBM's Centennial. 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. You can also follow him on Twitter @az990tony.
(Short URL for this blog: ibm.co/Pearson
( I cannot take credit for coining the new term "bleg". I saw this term firstused over on the [FreakonomicsBlog]. If you have not yet read the book "Freakonomics", I highly recommend it! The authors' blog is excellent as well.)
For this comparison, it is important to figure out how much workload a mainframe can support, how much an x86 cansupport, and then divide one from the other. Sounds simple enough, right? And what workload should you choose?IBM chose a business-oriented "data-intensive" workload using Oracle database. (If you wanted instead a scientific"compute-intensive" workload, consider an [IBM supercomputer] instead, the most recent of which clocked in over 1 quadrillion floating point operations per second, or PetaFLOP.) IBM compares the following two systems:
Sun Fire X2100 M2, model 1220 server (2-way)
IBM did not pick a wimpy machine to compare against. The model 1220 is the fastest in the series, with a 2.8Ghz x86-64 dual-core AMD Opteron processor, capable of running various levels of Solaris, Linux or Windows.In our case, we will use Oracle workloads running on Red Hat Enterprise Linux.All of the technical specifications are available at the[Sun Microsystems Sun Fire X1200] Web site.I am sure that there are comparable models from HP, Dell or even IBM that could have been used for this comparison.
IBM z10 Enterprise Class mainframe model E64 (64-way)
This machine can run a variety of operating systems also, including Red Hat Enterprise Linux (RHEL). The E64 has four "multiple processor modules" called"processor books" for a total of 77 processing units: 64 central processors, 11 system assist processors (SAP) and 2 spares. That's right, spare processors, in case any others gobad, IBM has got your back. You can designate a central processor in a variety of flavors. For running z/VM and Linux operating systems, the central processors can be put into "Integrated Facility for Linux" (IFL) mode.On IT Jungle, Timothy Patrick Morgan explains the z10 EC in his article[IBM Launches 64-Way z10 Enterprise Class Mainframe Behemoth]. For more information on the z10 EC, see the 110-page [Technical Introduction], orread the specifications on the[IBM z10 EC] Web site.
In a shop full of x86 servers, there are production servers, test and development servers, quality assuranceservers, standby idle servers for high availability, and so on. On average, these are only 10 percent utilized.For example, consider the following mix of servers:
125 Production machines running 70 percent busy
125 Backup machines running idle ready for active failover in case a production machine fails
1250 machines for test, development and quality assurance, running at 5 percent average utilization
While [some might question, dispute or challenge thisten percent] estimate, it matches the logic used to justify VMware, XEN, Virtual Iron or other virtualization technologies. Running 10 to 20 "virtual servers" on a single physical x86 machine assumes a similar 5-10 percent utilization rate.
Note: The following paragraphs have been revised per comments received.
Now the math. Jon, I want to make it clear I was not involved in writing the press release nor assisted with thesemath calculations. Please, don't shoot the messenger! Remember this cartoon where two scientists in white lab coats are writing mathcalculations on a chalkboard, and in the middle there is "and then a miracle happens..." to continue the rest ofthe calculations?
In this case, the miracle is the number that compares one server hardware platform to another. I am not going to bore people with details like the number of concurrent processor threads or the differencesbetween L1 and L3 cache. IBM used sophisticated tools and third party involvement that I am not allowed to talk about, and I have discussed this post with lawyers representing four (now five) different organizations already,so for the purposes of illustration and explanation only, I have reverse-engineered a new z10-to-Opteron conversion factor as 6.866 z10 EC MIPS per GHz of dual-core AMD Opteron for I/O-intensive workloads running only 10 percent average CPU utilization. Business applications that perform a lot of I/O don't use their CPU as much as other workloads.For compute-intensive or memory-intensive workloads, the conversion factor may be quite different, like 200 MIPS per GHz, as Jeff Savit from Sun Microsystems points out in the comments below.
Keep in mind that each processor is different, and we now have Intel, AMD, SPARC, PA-RISC and POWER (and others); 32-bit versus 64-bit; dual-core and quad-core; and different co-processor chip sets to worry about. AMD Opteron processors come in different speeds, but we are comparing against the 2.8GHz, so 1500 times 6.866 times 2.8 is 28,337. Since these would be running as Linux guestsunder z/VM, we add an additional 7 percent overhead or 2,019 MIPS. We then subtract 15 percent for "smoothing", whichis what happens when you consolidate workloads that have different peaks and valleys in workload, or 4,326 MIPS.The end is that we need a machine to do 26,530 MIPS. Thanks to advances in "Hypervisor" technological synergy between the z/VM operating system and the underlying z10 EC hardware, the mainframe can easily run 90 percent utilized when aggregating multiple workloads, so a 29,477 MIPS machine running at 90 percent utilization can handle these 26,530 MIPS.
N-way machines, from a little 2-way Sun Fire X2100 to the might 64-way z10 EC mainframe, are called "Symmetric Multiprocessors". All of the processors or cores are in play, but sometimes they have to taketurns, wait for exclusive access on a shared resource, such as cache or the bus. When your car is stopped at a red light, you are waiting for your turn to use the shared "intersection". As a result, you don't get linear improvement, but rather you get diminishing returns. This is known generically as the "SMP effect", and in IBM documentsthis as [Large System Performance Reference].While a 1-way z10 EC can handle 920 MIPS, the 64-way can only handle30,657 MIPS. The 29,477 MIPS needed for the Sun x2100 workload can be handled by a 61-way, giving you three extraprocessors to handle unexpected peaks in workload.
But are 1500 Linux guest images architecturally possible? A long time ago, David Boyes of[Sine Nomine Associates] ran 41,400 Linux guest images on a single mainframe using his [Test Plan Charlie], and IBM internallywas able to get 98,000 images, and in both cases these were on machines less powerful than the z10 EC. Neitherof these were tests ran I/O intensive workloads, but extreme limits are always worth testing. The 1500-to-1 reduction in IBM's press release is edge-of-the-envelope as well, so in production environments, several hundred guest images are probably more realistic, and still offer significant TCO savings.
The z10 EC can handle up to 60 LPARs, and each LPAR can run z/VM which acts much like VMware in allowing multipleLinux guests per z/VM instance. For 1500 Linux guests, you could have 25 guests each on 60 z/VM LPARs, or 250 guests on each of six z/VM LPARs, or 750 guests on two LPARs. with z/VM 5.3, each LPAR can support up to 256GB of memory and 32 processors, so you need at least two LPAR to use all 64 engines. Also, there are good reasons to have different guests under different z/VM LPARs, such as separating development/test from production workloads. If you had to re-IPLa specific z/VM LPAR, it could be done without impacting the workloads on other LPARs.
To access storage, IBM offers N-port ID Virtualization (NPIV). Without NPIV, two Linux guest images could not accessthe same LUN through the same FCP port because this would confuse the Host Bus Adapter (HBA), which IBM calls "FICON Express" cards. For example, Linux guest 1 asks to read LUN 587 block 32 and this is sent out a specific port, to a switch, to a disk system. Meanwhile, Linux guest 2 asks to read LUN 587 block 49. The data comes back to the z10 EC with the data, gives it to the correct z/VM LPAR, but then what? How does z/VM know which of the many Linux guests to give the data to? Both touched the same LUN, so it is unclear which made the request. To solve this, NPIV assigns a virtual "World Wide Port Name" (WWPN), up to 256 of them per physical port, so you can have up to 256 Linux guests sharing the same physical HBA port to access the same LUN.If you had 250 guests on each of six z/VM LPARs, and each LPAR had its own set of HBA ports, then all 1500 guestscould access the same LUN.
Yes, the z10 EC machines support Sysplex. The concept is confusing, but "Sysplex" in IBM terminology just means that you can have LPARs either on the same machine or on separate mainframes, all sharing the same time source, whether this be a "Sysplex Timer" or by using the "Server Time Protocol" (STP). The z10 EC can have STP over 6 Gbps Infiniband over distance. If you wantedto have all 1500 Linux guests time stamp data identically, all six z/VM LPARs need access to the shared time source. This can help in a re-do or roll-back situation for Oracle databases to complete or back-out "Units of Work" transactions. This time stamp is also used to form consistency groups in "z/OS Global Mirror", formerly called "XRC" for Extended Remote Distance Copy. Currently, the "timestamp" on I/O applies only to z/OS and Linux and not other operating systems. (The time stamp is done through the CDK driver on Linux, and contributed back to theopen source community so that it is available from both Novell SUSE and Red Hat distributions.)To have XRC have consistency between z/OS and Linux, the Linux guests would need to access native CKD volumes,rather than VM Minidisks or FCP-oriented LUNs.
Note: this is different than "Parallel Sysplex" which refers to having up to 32 z/OS images sharing a common "Coupling Facility" which acts as shared memory for applications. z/VM and Linux do not participate in"Parallel Sysplex".
As for the price, mainframes list for as little as "six figures" to as much as several million dollars, but I have no idea how much this particular model would cost. And, of course, this is just the hardware cost. I could not find the math for the $667 per server replacement you mentioned, so don't have details on that.You would need to purchase z/VM licenses, and possibly support contracts for Linux on System z to be fully comparable to all of the software license and support costs of the VMware, Solaris, Linux and/or Windows licenses you run on the x86 machines.
This is where a lot of the savings come from, as a lot of software is licensed "per processor" or "per core", and so software on 64 mainframe processors can be substantially less expensive than 1500 processors or 3000 cores.IBM does "eat its own cooking" in this case. IBM is consolidating 3900 one-application-each rack-mounted serversonto 30 mainframes, for a ratio of 130-to-1 and getting amazingly reduced TCO. The savings are in the followingareas:
Hardware infrastructure. It's not just servers, but racks, PDUs, etc. It turns out to be less expensive to incrementally add more CPU and storage to an existing mainframe than to add or replace older rack-em-and-stack-emwith newer models of the same.
Cables. Virtual servers can talk to each other in the same machine virtually, such as HiperSockets, eliminatingmany cables. NPIV allows many guests to share expensive cables to external devices.
Networking ports. Both LAN and SAN networking gear can be greatly reduced because fewer ports are needed.
Administration. We have Universities that can offer a guest image for every student without having a majorimpact to the sys-admins, as the students can do much of their administration remotely, without having physicalaccess to the machinery. Companies uses mainframe to host hundreds of virtual guests find reductions too!
Connectivity. Consolidating distributed servers in many locations to a mainframe in one location allows youto reduce connections to the outside world. Instead of sixteen OC3 lines for sixteen different data centers, you could have one big OC48 line instead to a single data center.
Software licenses. Licenses based on servers, cores or CPUs are reduced when you consolidate to the mainframe.
Floorspace. Generally, floorspace is not in short supply in the USA, but in other areas it can be an issue.
Power and Cooling. IBM has experienced significant reduction in power consumption and cooling requirementsin its own consolidation efforts.
All of the components of DFSMS (including DFP, DFHSM, DFDSS and DFRMM) were merged into a single product "DFSMS for z/OS" and is now an included element in the base z/OS operating system. As a result of these, customers typically have 80 to 90 percent utilization on their mainframe disk. For the 1500 Linux guests, however, most of the DFSMS features of z/OS do not apply. These functions were not "ported over" to z/VM nor Linux on any platform.
Instead, the DFSMS concepts have been re-implemented into a new product called "Scale-Out File Services" (SOFS) which would provide NAS interfaces to a blendeddisk-and-tape environment. The SOFS disk can be kept at 90 percent utilization because policies can place data, movedata and even expire files, just like DFSMS does for z/OS data sets. SOFS supports standard NAS protocols such as CIFS,NFS, FTP and HTTP, and these could be access from the 1500 Linux guests over an Ethernet Network Interface Card (NIC), which IBM calls "OSA Express" cards.
Lastly, IBM z10 EC is not emulating x86 or x86-64 interfaces for any of these workloads. No doubt IBM and AMD could collaborate together to come up with an AMD Opteron emulator for the S/390 chipset, and load Windows 2003 right on top of it, but that would just result in all kinds of emulation overhead.Instead, Linux on System z guests can run comparable workloads. There are many Linux applications that are functionally equivalent or the same as their Windows counterparts. If you run Oracle on Windows, you could runOracle on Linux. If you run MS Exchange on Windows, you could run Bynari on Linux and let all of your Outlook Expressusers not even know their Exchange server had been moved! Linux guest images can be application servers, web servers, database servers, network infrastructure servers, file servers, firewall, DNS, and so on. For nearly any business workload you can assign to an x86 server in a datacenter, there is likely an option for Linux on System z.
Hope this answers all of your questions, Jon. These were estimates based on basic assumptions. This is not to imply that IBM z10 EC and VMware are the only technologies that help in this area, you can certainly find virtualization on other systems and through other software.I have asked IBM to make public the "TCO framework" that sheds more light on this.As they say, "Your mileage may vary."
For more on this series, check out the following posts:
If in your travels, Jon, you run into someone interested to see how IBM could help consolidate rack-mounted servers over to a z10 EC mainframe, have them ask IBM for a "Scorpion study". That is the name of the assessment that evaluates a specific clientsituation, and can then recommend a more accurate estimate configuration.
I am still wiping the coffee off my computer screen, inadvertently sprayed when I took a sip while reading HDS' uber-blogger Hu Yoshida's post on storage virtualization and vendor lock-in.
HDS is a major vendor for disk storage virtualization, and Hu Yoshida has been around for a while, so I felt it was fair to disagree with some of the generalizations he made to set the record straight. He's been more careful ever since.
However, his latest post [The Greening of IT: Oxymoron or Journey to a New Reality] mentions an expert panel at SNW that includedMark O’Gara Vice President of Infrastructure Management at Highmark. I was not at the SNW conference last week in Orlando, so I will just give the excerpt from Hu's account of what happened:
"Later I had the opportunity to have lunch with Mark O’Gara. Mark is a West Point graduate so he takes a very disciplined approach to addressing the greening of IT. He emphasized the need for measurements and setting targets. When he started out he did an analysis of power consumption based on vendor specifications and came up with a number of 513 KW for his data center infrastructure....
The physical measurements showed that the biggest consumers of power were in order: Business Intelligence Servers, SAN Storage, Robotic tape Library, and Virtual tape servers....
Another surprise may be that tape libraries are such large consumers of power. Since tape is not spinning most of the time they should consume much less power than spinning disk - right? Apparently not if they are sitting in a robotic tape library with a lot of mechanical moving parts and tape drives that have to accelerate and decelerate at tremendous speeds. A Virtual Tape Library with de-duplication factor of 25:1 and large capacity disks may draw significantly less power than a robotic tape library for a given amount of capacity.
Obviously, I know better than to sip coffee whenever reading Hu's blog. I am down here in South America this week, the coffee is very hot and very delicious, so I am glad I didn't waste any on my laptop screen this time, especially reading that last sentence!
In that report, a 5-year comparison found that a repository based on SATA disk was 23 times more expensive overall, and consumed 290 times more energy, than a tape library based on LTO-4 tape technology. The analysts even considered a disk-based Virtual Tape Library (VTL). Focusing just on backups, at a 20:1 deduplication ratio, the VTL solution was still 5 times per expensive than the tape library. If you use the 25:1 ratio that Hu Yoshida mentions in his post above, that would still be 4 times more than a tape library.
I am not disputing Mark O'Gara's disciplined approach. It is possible that Highmark is using a poorly written backup program, taking full backups every day, to an older non-IBM tape library, in a manner that causes no end of activity to the poor tape robotics inside. But rather than changing over to a VTL, perhaps Mark might be better off investigating the use of IBM Tivoli Storage Manager, using progressive backup techniques, appropriate policies, parameters and settings, to a more energy-efficient IBM tape library.In well tuned backup workloads, the robotics are not very busy. The robot mounts the tape, and then the backup runs for a long time filling up that tape, all the meanwhile the robot is idle waiting for another request.
(Update: My apologies to Mark and his colleagues at Highmark. The above paragraph implied that Mark was using badproducts or configured them incorrectly, and was inappropriate. Mark, my full apology [here])
If you do decide to go with a Virtual Tape Library, for reasons other than energy consumption, doesn't it make sense to buy it from a vendor that understands tape systems, rather than buying it from one that focuses on disk systems? Tape system vendors like IBM, HP or Sun understand tape workloads as well as related backup and archive software, and can provide better guidance and recommendations based on years of experience. Asking advice abouttape systems, including Virtual Tape Libraries, from a disk vendor is like asking for advice on different types of bread from your butcher, or advice about various cuts of meat at the bakery.
The butchers and bakers might give you answers, but it may not be the best advice.
I am proud to announce that fellow IBMer Carlos Pratt has launched a new IBM storage blog[GreenSpeed].
I'd like to expand a bit on how I know Carlos. Back in 1999 I was asked to lead a team at IBM Tucson to install Linux on our local z800 mainframe, and run tests to confirm that all of our IBM disk and tape storage offerings attached successfully. I was, at the time, lead architect for DFSMS on OS/390 and management felt that my knowledge of the S/390 instruction set was all that was needed to pull this off. My team was a collection of people from a variety of other hardware and software teams, and Carlos came over from the Disk Performance test team.
Needless to say, there were some challenges. The port of Red Hat and SUSE Linux over to the mainframe required special device drivers, and in some cases, we actually needed to make changes to the Linux kernel. While it was over 100 degrees outside, we were in the test lab wearing jackets with a refrigerator thermometer hanging on the wall to monitor our ice cold working conditions.
And of course, we had our internal skeptics. At the time, Linux was only a few percentage points of marketshare, and a few unenlightened souls did not see any reason to invest in support for a new operating system until it was more established. People with a "Wait-and-See" attitude don't last long at IBM. Fortunately, smarter heads prevailed, and now that Linux is well established as the operating system of the future, we can all look back and say "I told you so!"
Carlos was a "get things done" kind of guy. Working with frequent patches to the Linux kernel, device drivers under development, and a team fairly new to this new operating system, Carlos was able to provide the driving force to get our tests done.
I’ve just returned from the IBM Tivoli Pulse conference in Las Vegas – a meeting of over 4000 customers, partners, and IBM employees. ... There was a lot to digest, but three of the major themes caught my attention, and my imagination. ... First, IBM put a huge push behind their Dynamic Infrastructure initiative. Sounds like so many other automation and autonomic initiatives of the past, right? Well, things are getting better, and “dynamic” is becoming more of a realistic possibility, especially with the emergence of cloud computing and cloud services models. ... Second, a lot of time was spent on IBM’s Service Management Industry Solutions. When I first heard of this, my thought was that IBM was creating solutions for the Service Management industry (i.e. food services, janitorial services, hospitality services). But this is much larger than that – much, much larger. IBM is taking their unique ability to pair business (non-IT) expertise with IT consulting, planning, and technology delivery, and constructing (careful – here comes the “f” word) frameworks for several vertical industry segments. ... IBM is perhaps the only organization in the world that can take this on fully and hope to deliver a meaningful result. But beyond that, this represents a huge opportunity for IT professionals to become the transformation agents within their own organizations, contributing at a whole new level. ... Lastly, I was really impressed by IBM’s Smarter Planet initiative. The primary thought here was that the key to a greener planet is to take inefficiencies out of just about every form of business through the intelligent application and deployment of technology. At first I was thinking this was just another marketing initiative, but in the course of this event, listening to the keynotes and talking to a number of IBM execs, it became apparent that this is a substantial cultural shift within IBM itself. Just think about that for a moment – when 400,000 employees all change their direction and focus, their sheer mass is going to make a noticeable difference. ... Magic (Johnson) gave an excellent talk, and reminded the audience that you should do two things no matter what your job or role. First, service starts with knowing your customers – not just who they are, but what they do and what is important to them. And second – always over-deliver. Go that extra step. Exceed expectations. The boost in loyalty, goodwill, and improved customer relationships will be well worth the effort. Good thoughts to keep with us….
If you missed Pulse 2009, perhaps because your company has put a clamp down on travel expenses, you are in luck! IBM is hosting the "Dynamic Infrastructure Forum" March 3-4, 2009, on your computer. This is an IBM Virtual event, no travel required! [Register Today!]
Last week, a writer for a magazine contacted us at IBM to confirm a quote that writing a Terabyte (TB) on disk saves 50,000 trees. I explained that this was cited from UC Berkeley's famousHow Much Information? 2003 study.
To be fair, the USA Today article explains that AT&T also offers "summary billing" as well as "on-line billing", but apparently neither of these are the default choice. I can understand that phone companies send out bills on paper because not everyone who has a phone has internet access, but in the case of its iPhone customers, internet access is in the palm of your hands! Since all iPhone customers have internet access, and AT&T knows which customers are using an iPhone, it would make sense for either on-line billing or summary billing to be the default choice, and let only those that hate trees explicitly request the full billing option.
Sending a box of 300 pages of printed paper is expensive, both for the sender and the recipient. This informationcould have been shipped less expensively on computer media, a single floppy diskette or CDrom for example. Forthose who prefer getting this level of detail, a searchable digitized version might be more useful to the consumer.
Which brings me to the concept of Information Lifecycle Management (ILM). You can read my recent posts on ILM byclicking the Lifecycle tab on the right panel, or my now infamous post from last year about ILM for my iPod.
His recollection of the history and evolution of ILM fairly matches mine:
The phrase "Information Lifecycle Management" was originally coined by StorageTek in early 1990s as a way to sell its tape systems into mainframe environments. Automated tape libraries eliminated most if not all of the concerns that disk-only vendors tout as the problem with manual tape. I began my IBM career in a product now called DFSMShsm which specifically moved data from disk to tape when it no longer needed the service level of disk. IBM had been delivering ILM offerings since the 1970s, so while StorageTek can't claim inventing the concept, we give them credit for giving it a catchy phrase.
EMC then started using the phrase four years ago in its marketing to sell its disk systems, including slower less-expensive SATA disk. The ILM concept helped EMC provide context for the many acquisitions of smaller companies that filled gaps in the EMC portfolio. Question: Why did EMC acquire company X? Answer: To be more like IBM and broaden its ILM solution portfolio.
Information Lifecycle Management is comprised of the policies, processes,practices, and tools used to align the business value of information with the mostappropriate and cost effective IT infrastructure from the time information isconceived through its final disposition. Information is aligned with businessrequirements through management policies and service levels associated withapplications, metadata, and data.
Whitepapers and other materials you might read from IBM, EMC, Sun/StorageTek, HP and others will all pretty much tell you what ILM is, consistent with this SNIA definition, why it is good for most companies, and how it is not just about buying disk and tape hardware. Software, services, and some discipline are needed to complete the implementation.
While the SNIA definition provides a vendor-independent platform to start the conversation, it can be intimidatingto some, and is difficult to memorize word for word.When I am briefing clients, especially high-level executives, they often ask for ILM to be explained in simpler terms. My simplified version is:
Information starts its life captured or entered as an "asset" ...
This asset can sometimes provide competitive advantage, or is just something needed for daily operations. Digital assets vary in business value in much the same way that other physical assets for a company might. Some assets might be declared a "necessary evil" like laptops, but are tracked to the n'th degree to ensure they are not lost, stolen or taken out of the building. Other assetsare declared "strategically important" but are readily discarded, or at least allowed to walk out the door each evening.
... then transitions into becoming just an "expense" ...
After 30-60 days, many of the pieces of information are kept around for a variety of reasons. However, if it isn'tneeded for daily operations, you might save some money moving it to less expensive storage media, throughless expensive SAN or LAN network gear, via less expensive host application servers. If you don't need instantaccess, then perhaps the 30 seconds or so to fetch it from much-less-expensive tape in an automated tape librarycould be a reasonable business trade-off.
... and ends up as a "liability".
Keeping data around too long can be a problem. In some cases, incriminating, and in other cases, just having toomuch data clogs up your datacenter arteries. If not handled properly within privacy guidelines, data potentially exposes sensitive personal or financial information of your employees and clients. Most regulations require certain data to be kept, in a manner protected against unexpected loss, unethical tampering, and unauthorized access, for a specific amount of time, after which it can be destroyed, deleted or shredded.
So ILM is not just a good idea to save a company money, it can keep them out of the court room, as well as help save the environment and not kill so many trees. Now that 100 percent of iPhone customers have internet access, and a goodnumber of non-iPhone customers have internet access at home, work, school or public library, it makes sense for companies to ask people to "opt-in" to getting their statements on paper, rather than forcing them to "opt-out".
An avid reader of this blog pointed me to a blog post [A Small Tec DIGG on IBM XIV], byGowri Ananthan, a System Engineer in Singapore.Basically, she covers past battles, er.. discussions between me and fellow blogger BarryB from EMC, and [blegs] foranswers to three questions.
Gowri, here are your answers:
Q1. Does IBM offer a Pay-as-you-Go [PAYGO] upgrade path for its IBM XIV disk storage system?
The concern was expressed as:
PAYGO also requires the customer to purchase the remaining capacity within 12 months of installation. So it is More of a 12-month installment plan than pay-as-you-grow.
A1. Actually, IBM offers several methods for your convenience:
With IBM's Capacity on Demand (CoD) plan, you get the full framewith 15 modules installed on your data center floor, but only pay for the first four modules 21 TB, then pay for 5.3TB module increments as you need them over the next 12 months. This is ideal for companies that don't know how fast they will grow, but do not want to wait for new modules to be delivered and installed when needed.
With IBM's Partial Rack offering, you can get a system with as little as six modules (27TB),and then over time, add more modules as you need. This does not have to be done within 12 months, you can stay at six modules for as long as you like, and you can take as long asyou want to add more modules. When you are ready for more capacity, the drawer or drawerscan be delivered, and installed non-disruptively.
Neither of these are "payment installment plans", but certainly if you want to spread yourcosts into regularly-scheduled monthlypayments across multiple years, IBM Global Financing can probably work something out.
Q2. Does IBM consider the XIV as green storage?
The concern was expressed as:
You are powering (8.4KW) and cooling all 180 drives for the whole duration, whether you're using the capacity or not. is it what you called Greener power usage..?
A2. Yes. IBM considers the IBM XIV as green storage. The 8.4KW per frame is lessthan the 10-plus KW that a comparable 2-frame EMC DMX-950 system would consume. Theenergy savings in IBM XIV comes from delivering FC-like speeds using slower SATA disks that rotate slower, and therefore take less energy to spin.
In the fully-populated or Capacity on Demand configuration, you would spin all 180disks. However, using the partial rack configuration, the 6-module has only 40 percent ofthe disks, and therefore consumes only 40 percent of the energy. If you don't plan to storeat least 20-30 TB, you might consider the DS3000, DS4000, DS5000, or DS8000 disk system instead.
Q3. How do you connect more than 24 host ports to an IBM XIV?
The concern was expressed as:
And finally do not forget my question on 24-FC Ports… Up to 24 Fiber Channel ports offering 4 Gbps, 2Gbps or 1 Gbps multi-mode and single-mode support.Stop.. stop.. how you gonna squeeze existing bunch of FC cables in 24 ports?
A3. Best practices suggest that if you have ten or more physical servers, each with two separate FC ports, then you should use a SAN switch or director in between. If you require four ports per server, then you would need a SAN switch beyond six servers to connect to the IBM XIV. If you consider that 24 FC ports, at 4Gbps, represents nearly 10 GB/sec of bandwidth, you will recognize that this is not a performance bottleneck for the system.
This week I'm in beautiful Guadalajara, Mexico teaching at our[System Storage Portfolio Top Gun class].We have all of our various routes-to-market represented here, including our direct sales force, our technicalteams, our online IBM.COM website sales, as well as IBM Business Partners.Everyone is excited over last week's IBM announcement of [4Q07 and full year 2007 results], which includesdouble-digit growth in our IBM System Storage business, led by sales of our DS8000, SAN Volume Controller and Tapesystems. Obviously, as an IBM employee and stockholder, I am biased, so instead I thought I would provide someexcerpts from other bloggers and journalists.
But what was striking in the company’s conference call on Thursday afternoon was the unhedged optimism in its outlook for 2008, given the strong whiff of recession fear elsewhere.
The questions from Wall Street analysts in the conference call had a common theme. Why are you so comfortable about the 2008 outlook? Now, that might just be professional churlishness, since so many of them have been so wrong recently about I.B.M. Wall Street had understandably thought, for example, that I.B.M.’s sales to financial services companies — the technology giant’s largest single customer category — would suffer in the fourth quarter, given the way banks have been battered by the mortgage credit crunch.
But Mr. Loughridge said that revenue from financial services customers rose 11 percent in the fourth quarter, to $8 billion. The United States, he noted, accounts for only 25 percent of I.B.M.’s financial services business.
The other thing that seems apparent is how much I.B.M.’s long-term strategy of moving up to higher-profit businesses and increasingly relying on services and software is working. Its huge services business grew 17 percent to $14.9 billion in the quarter. After the currency benefit, the gain was 10 percent, but still impressive. Software sales rose 12 percent to $6.3 billion.
Looking at IBM's business segments, it can be seen that they offer far more coverage of the technology space that those of the typical tech company:
IBM is just so big and diversified that there is little comparison between it and most other tech companies. IBM is a member of an elite group of companies like Cisco Systems (CSCO), Microsoft (MSFT), Oracle (ORCL) or Hewlett-Packard (HPQ).
IBM's wide international coverage and deep technological capabilities dwarf those of most tech companies. Not only do they have sales organizations worldwide but they have developers, consultants, R&D workers and supply chain workers in each geographic region. Their product mix runs from custom software to packaged enterprise software, hardware (mainframes and servers), semiconductors, databases, middleware technology, etc., etc. There are few tech companies that even attempt to support that many kinds and variations of products.
As color on the fourth quarter earnings announcement, there are a couple of observations that I would like to make. The first one speaks to IBM's international prowess. The company indicated that growth in the Americas was only 5%. International sales were a primary driver of IBM's good results. As an insight on the difference between IBM and most other tech companies, it is clear that nowadays, a tech company that isn't adept at selling internationally is going to be in trouble.
Terrific performance in a terrific year - no doubt a result of its strong global model. IBM operates in 170 countries, with about 65% of its employees outside US and about 30% in Asia Pacific. For fiscal 2007, revenues from Americas grew 4% to $41.1 billion (42% of total revenue), [EMEA] grew 14% to $34.7 billion (35%of total revenue), and Asia-Pacific grew by 11% to $19.5 billion (19.7% of total revenue). IBM sees growth prospects not just in [BRIC] but also countries like Malaysia, Poland, South Africa, Peru, and Singapore.
Thus far 2008–all two weeks of it–hasn’t been a pretty for the tech industry. Worries about the economy prevail. And even companies that had relatively good things to say like Intel get clobbered. It’s ugly out there–unless you’re IBM.
I am sure there will be more write-ups and analyses on this over the next coming weeks, and others will probably waituntil more tech companies announce their results for comparison.
Last week, I presented IBM's strategic initiative, the IBM Information Infrastructure, which is part of IBM's New Enterprise Data Center vision. This week, I will try to get around to talking about some of theproducts that support those solutions.
I was going to set the record straight on a variety of misunderstandings, rumors or speculations, but I think most have been taken care of already. IBM blogger BarryW covered the fact that SVC now supports XIV storage systems, in his post[SVC and XIV],and addressed some of the FUD already. Here was my list:
Now that IBM has an IBM-branded model of XIV, IBM will discontinue (insert another product here)
I had seen speculation that XIV meant the demise of the N series, the DS8000 or IBM's partnership with LSI.However, the launch reminded people that IBM announced a new release of DS8000 features, new models of N series N6000,and the new DS5000 disk, so that squashes those rumors.
IBM XIV is a (insert tier level here) product
While there seems to be no industry-standard or agreement for what a tier-1, tier-2 or tier-3 disk system is, there seemed to be a lot of argument over what pigeon-hole category to put IBM XIV in. No question many people want tier-1 performance and functionality at tier-2 prices, and perhaps IBM XIV is a good step at giving them this. In some circles, tier-1 means support for System z mainframes. The XIV does not have traditional z/OS CKD volume support, but Linux on System z partitions or guests can attach to XIV via SAN Volume Controller (SVC), or through NFS protocol as part of the Scale-Out File Services (SoFS) implementation.
Whenever any radicalgame-changing technology comes along, competitors with last century's products and architectures want to frame the discussion that it is just yet another storage system. IBM plans to update its Disk Magic and otherplanning/modeling tools to help people determine which workloads would be a good fit with XIV.
IBM XIV lacks (insert missing feature here) in the current release
I am glad to see that the accusations that XIV had unprotected, unmirrored cache were retracted. XIV mirrors all writes in the cache of two separate modules, with ECC protection. XIV allows concurrent code loadfor bug fixes to the software. XIV offers many of the features that people enjoy in other disksystems, such as thin provisioning, writeable snapshots, remote disk mirroring, and so on.IBM XIV can be part of a bigger solution, either through SVC, SoFS or GMAS that provide thebusiness value customers are looking for.
IBM XIV uses (insert block mirroring here) and is not as efficient for capacity utilization
It is interesting that this came from a competitor that still recommends RAID-1 or RAID-10 for itsCLARiiON and DMX products.On the IBM XIV, each 1MB chunk is written on two different disks in different modules. When disks wereexpensive, how much usable space for a given set of HDD was worthy of argument. Today, we sell you abig black box, with 79TB usable, for (insert dollar figure here). For those who feel 79TB istoo big to swallow all at once, IBM offers "capacity on demand" pricing, where you can pay initially for as littleas 22TB, but get all the performance, usability, functionality and advanced availability of the full box.
IBM XIV consumes (insert number of Watts here) of energy
For every disk system, a portion of the energy is consumed by the number of hard disk drives (HDD) andthe remainder to UPS, power conversion, processors and cache memory consumption. Again, the XIV is a bigblack box, and you can compare the 8.4 KW of this high-performance, low-cost storage one-frame system with thewattage consumed by competitive two-frame (sometimes called two-bay) systems, if you are willing to take some trade-offs. To getcomparable performance and hot-spot avoidance, competitors may need to over-provision or use faster, energy-consuming FC drives, and offer additional software to monitor and re-balance workloads across RAID ranks.To get comparable availability, competitors may need to drop from RAID-5 down to either RAID-1 or RAID-6.To get comparable usability, competitors may need more storage infrastructure management software to hide theinherent complexity of their multi-RAID design.
Of course, if energy consumption is a major concern for you, XIV can be part of IBM's many blended disk-and-tapesolutions. When it comes to being green, you can't get any greener storage than tape! Blended disk-and-tapesolutions help get the best of both worlds.
Well, I am glad I could help set the record straight. Let me know what other products people you would like me to focus on next.
Yesterday marked the first day of Spring here in the Northern hemisphere, and often this means it is timefor some "Spring cleaning". This is a great time to re-evaluate all of your stuff and clean house.
In the bits-vs-atoms discussion, Annie Leonard has a quick [20-minute video] about the atoms side of stuff,from extraction of natural resources, production, distribution, consumption, to final disposal.
On the bits side of things, the picture is much different.
We don't really extract information,rather we capture it, and lately that process is done directly into digital formats, from digital photography, digital recording of music, and so on. A lot of medical equipmentnow take X-rays and other medical images directly into digital format. By 2011, it is estimated that as much as 30 percent of all storage will be for holding medical images.
Production refers to the process of combining raw materials and making them into something useful. The sameapplies to information, there are a variety of ways to make information more presentable. In the Web 2.0 world, these are called Mashups, combiningraw information in a manner that are more usable.Fellow IBM blogger Bob Sutor discusses IBM's latest contribution, SMash, in his post[Secure Mashups via SMash].
According to Tim Sanders, 90 percent of business information is distributed by email, but less than 10 percentof employees are formally trained to distribute information correctly. Here's a quick 3-minute trailerto his "Dirty Dozen" rules of how to do email properly.
I have not watched the DVD that this trailer is promoting, but I certainly agree with the overall concept.
This week I also had the pleasure to hear [Art Mortell], author ofthe book The Courage to Fail: Art Mortell's Secrets to Business Success. He gave an inspirational talk about how to deal with our stressful lives. One key pointwas that stress often came from our own expectations. This is certainly true on how we consume information.Often times our expectations determine how well we read, watch or listen to information being presented.Sometimes information is factually correct, but presented in such a boring manner that it is just toodifficult to consume.
John Windsor on YouBlog takes this one step further, asking [Are you predictable?]He makes a strong case on why presenting in a predictable manner can actually hurt your chances of communication.
And finally, there is disposal. We are all a bunch of digital pack-rats. With atoms, you eventuallyrun out of closet space, with bits the problem is not as obvious, and often can be resolved by spendingyour way out of it. On average, companies are expanding their storage capacity by 57 percent every year. Thatworked well when dollar-per-GB prices of disk dropped to match, but now technology advancements are slowing down. Diskwill not be dropping in price as fast as you need, and now might be a good time to re-evaluate your"Keep everything forever" strategy.
Consider "Spring cleaning" to be an excellent excuse to evaluate the data you have on your disk systems.Should it be on disk? Will it be accessed often enough to justify that cost? Does it need immediateonline access times, or can waiting a minute or two for a tape mount from an automated library be sufficient?Does it represent business value?
I have been to customers that have discovered a lot of "orphan data" on their disk systems. This isdata that does not belong to anyone currently working at the company. Maybe the owners of the data retired,were laid off, or even fired, but nobody bothered to clean up their files after they left the company.
I've also seen a lot of "stale data" on disk, data that has not be read or written in the past 90 days.Are you spending 13-18 watts of energy to spin each disk drive just to contain data nobody ever looks at?
In some cases, orphan or stale data represents business value, and need to be kept around for businessor legal reasons. Perhaps some government regulation requires you to retain this information for someyears. In that case, rather than deleting it, move it to tape, perhaps using theIBM System Storage DR550 to protect it for the time required and handle its eventual disposal.
Certainly something to think about, while you snap the ears off those chocolate bunnies, watching yourkids run around looking for eggs. Enjoy your weekend!
It's been a while since I've talked about [Second Life].
The latest post on eightbar[Spimes, Motes and Data centers]discusses IBM's use of virtual world technology to analyze data centers in three dimensions.New World Note asks[What's The Point Of 3D Data Centers?]One would think that a simple monitoring tool based on a two-dimensional floor plan would be enough to evaluate a data center.
Enter Michael Osias, IBM (a.k.a Illuminous Beltran in Second Life). Some of the leading news sites havebegun to notice some 3D data centers that he has helped pioneer. UgoTrade writes up an article aboutMichael and the media attention in [The Wizard of IBM's 3DData Centers].
Of course, in presenting these "Real Life/Second Life" (RL/SL) interactive technologies, IBM is sometimes the target of ridicule. Why? Because IBM is 10 years ahead of everyone else. So, are there aspects of a data center where 3D interfaces makes sense? I think there is.
IBM TotalStorage Productivity Center has an awesome "topology viewer" that shows what servers are connectedto which switches, to which disk systems and tape libraries. This is all done in a 2D diagram, generated dynamicallywith data discovered through open standard interfaces, similar to what you might draw manually with toolslike Visio. Imagine, however, howmore powerful if it were a 3D viewer, with virtual equipment mapped to the physical location of each pieceof hardware on the data center floor, including the position on the rack and location on the data center floor.
Designing computer room air conditioning (CRAC) systems is actually a three dimensional problem. Cold air isfed underneath the raised floor, comes up through strategically placed "vent" tiles, taken in the front ofeach rack. Hot air comes out the back of each rack, and hopefully finds ceiling duct intake to get cooled again.The temperature six inches off the floor is different than the temperature six feet off the floor, and 3Dmonitor tools could be helpful in identifying "hot spots" that need attention. In this case "spimes" representsensors in the 3D virtual world, able to report back information to help diagnose problems or monitor events.
After many people left the mainframe in favor of running a single application per distributed server, the pendulumhas finally swung back. Companies are discovering the many benefits of changing this behavior. "Re-centralization" is the task at hand. Thanks to virtualization of servers, networks and storage, sharing common resources canonce again claim the benefits of economies of scale. In many cases, servers work together in collective unitsfor specific applications that might benefit better if consolidated together onto the same equipment.
IBM's "New Enterprise Data Center" vision recognizes that people will need to focus on the management aspectsof their IT infrastructure, and 3D virtual world technologies might be an effective way to getthe job done.
You could buy 10 liters of gasoline in Venezuela with this coin.
I'm back from South America, and am now in Chicago, Illinois. I'm having breakfast at the Starbucksdowntown, and thought I would make a post before all of my meetings today.
On this trip, I met with IBM Business Partners and sales reps from Argentina, Colombia, Ecuador and Venezuela. While I have visited thefirst three countries on past trips, this was my first time to Caracas, Venezuela. I grew up in La Paz, Bolivia, and speak Spanish fluently, so had no problemgetting around and holding discussions with everyone. While my friends in the US are oftensurprised I speak multiple languages, it doesn't surprise anyone I visit in other countries.If you are going to have worldwide job responsibilities for a global company that does businessin over 180 countries, the least you could do is learn a few additional languages. I suspect themajority of the 350,000 IBM employees speak at least two languages, the exceptions being mostly the 50,000 orso employees that live in the United States.
I flew on American Airlines from Tucson to Dallas to Caracas, and was only slightly delayed as a resultof all of the flight cancellations that happened earlier that week. Some companies designate a single "official airline" for their employees to use. That makessense if all of your employees are located in a single city, and that city is the hub for yourdesignated airline.IBM is too big, too spread out, and sells technology to nearly every airline to make sucha designation. Instead, IBM tries to spread its business out to multiple carriers, although all ofmy colleagues seems to have their own personal favorites. Mine are American Airlines, Singapore Airlines and Cathay Pacific.
While other people were upset over the delays, I found American Airlines did a great job keeping me informed,and all their employees I talked to seemed to be handling the situation fairly well. If youfly on American, I recommend you sign up for "text message" notifications. I did this for everyleg of my trip, and was kept up to date on times, gates and status. Very helpful!American Airlines even started their own corporate blog: [AA Conversation] (Special thanks to my friend[Paul Gillen] for pointing this out)
(I read somewhere that if you are going to travel anywhere, you need to remember to bringboth your sunscreen and your sense of humor, otherwise you are going to get burned. Goodadvice! Trust me, you don't even know how bad it can really be until you travel in the third world.)
Anyhoo, last week, IBM Venezuela celebrated its 70th anniversary. That's right, IBM has been doingbusiness in Venezuela for the past 70 years. Also last week, IBM put out its impressive [1Q08 quarterly results],including 10 percent growth for IBM System Storage product line worldwide, comparing what IBM earned this first quarter to what IBM earned the first quarter of last year. For just the Latin American countries,the growth for IBM System Storage was 20 percent!There are a lot of oil and gas companies in Venezuela. With a barrel of oil selling at more than$117 US dollars, these companies are looking to spend their newly earned profits on IBM systems, software and services.
As for the picture above, that is a one-thousand Bolivares coin, worth about 47 US cents atthis week's official exchange rate. As with many Latin American countries going through [years of high inflation], Venezuela was tired of all those zeros on their money. For example, a cheeseburger, freedom fries and a Cokeat McDonald's would set you back 20,000 Bolivares.This year the Venezuelan governmentcreated a new currency called "Bolivares Fuertes" (VEF), lopping off the last three zeros.So, the coin above would be replaced by a new coin with a big "1" on it instead, and an old 2000 Bolivares billwould be replaced by a new 2 Bolivares Fuertes bill. Unfortunately,I had to give all my new Venezuelan money back at the airport upon leaving, but they let me keep the coinabove, since it is old money, as a souvenir so that I could use it as a ball mark for playing golf.
(The term Bolivares is named after Simon Bolivar who was born in Caracas. He is famous throughoutSouth America, and was, and I am not making this up, the first president of Colombia, the secondpresident of Venezuela, the first president of Bolivia, and the sixth president of Peru. Here isthe [Wikipedia article] to learn more.)
Gasoline costs a mere 100 old Bolivares per liter.For those who don't do metric, gasoline therefore costsless than 18 cents per gallon. By comparison, in the USA, the average today was $3.47 US dollarsper gallon, of which 18.4 cents of this is Federal tax. That's right, we pay more just in taxes forgasoline than los venezolanos pay for it all.
The side effect of cheap gas is bad traffic. Everybody in Venezuela drives their own car, and nobody thinksabout the price of gasoline, carpooling, or taking public transportation, acting much like Americans used to, up until a few years ago. With some of the gridlock we faced, it might have been faster (but not safer)to walk there instead.
Which makes me wonder if American Airlines fills up their airplanes with fuel at these lower prices when theypick up people in Caracas to take them back to the United States. In 2002, fuel represented 10 percentof the average airline's operating expenses, but today it is now 25 percent. That is a drastic increase!
The same is happening in data centers. In the past, electricity was so cheap, and such a small percentof the total IT budget, nobody gave it much thought. But as the usage of electricity increased, andthe cost per KWh went up, this has a multiplying effect, and the growth in power and cooling costs isgrowing four times faster than the average IT hardware budget increase.
Well, this has been an interesting two weeks. On week 1, I focused on IBM's strategy and four keysolutions areas: Information Availability, Information Security, Information Retention, and InformationCompliance. On week 2, I focused on individual products, their attributes, features and functions.Which week drew more blog traffic? You guessed it--week 1. Apparently, people want to know more aboutsolutions to their challenges and problems, and not just see what piece part components are available.
While IBM had switched over to solution-selling a while ago, some of our competitors are still inproduct-selling mode, and try to frame all competitive comparisons on a product-by-product basis.In my post[Supermarkets and Specialty Shops], I drew the analogy that the IT supermarkets (IBM, HP, Sun and Dell) are focusedon selling solutions, but the IT specialty shops (HDS, EMC, and others) are still focused on products.
Certainly, the transition from product-focused to solution-focused is not an easy one. As the IT industry matures, more and more clients are looking to buy solutions from theirvendors. What does it take to change behaviour of newly acquired employees, recently hired sales reps, and business partners, many of whom come from product-centric cultures, to match this dramatic shift in the marketplace? Let's take a look at change in other areas of the world.
On the[Freakonomics blog], Stephen Dubner discusses how clever people in Israel have figured out a way to get people to clean up after their pets in public places. This is a problem in many countries. Here we see an old idea, the [carrot-and-stick] approach, combined with newinformation technology. Here's an excerpt:
"In order to keep a city’s streets clean of dog poop, require dog owners to submit DNA samples from their pets when they get licenses; then use that DNA database to trace any left-behind poop and send the dogs’ owners stiff fines.
Well, it took three years but the Israeli city of Petah Tikva has actually put this plan to work:
The city will use the DNA database it is building to match feces to a registered dog and identify its owner.
Owners who scoop up their dogs’ droppings and place them in specially marked bins on Petah Tikva’s streets will be eligible for rewards of pet food coupons and dog toys.
But droppings found underfoot in the street and matched through the DNA database to a registered pet could earn its owner a municipal fine."
Sometimes, if enough people change, then changing behaviours of the few remaining becomes much easier. DanLockton on his Architectures of Control blog posts about the[London Design Festival - Greengaged]. This year, the festival focused on behavior changes for a greener environment, ecodesign and sustainable issues in design.Here's an excerpt and corresponding 5-minute YouTube video:
Lea argued three important points relevant to behaviour change:
Behaviour change requires behaviour (i.e. the behaviour of others: social effects are critical, as we respond to others’ behaviour which in turn affects our own; targeting the ‘right’ people allows behaviour to spread)
Behaviour and motivation are two different things: To change behaviour, you need to understand and work with people’s motivations - which may be very different for different people.
Desire is not enough: lots of people desire to behave differently, but it needs to be very easy for them to do it before it actually happens."
Of course, tax and government regulations can heavily influence behaviour and decisions. Since today is[International Talk Like a Pirate Day], I thought I would finish this post off with this interesting piece on Google barges. Some companies, like IBM and Google, seem more adaptable to changing behaviour and trying out fresh new ideas.Will Runyon over on the Raised Floor blog, has a post about Google's patent for[Data center barges on the sea]:"The idea is to use waves to power the data centers, ocean water to cool them, and a moored distance of seven miles or more to avoid paying taxes."
Arrr! Now that's what I call a new way of looking at things!
Over the past year and a half, I have been focused on explaining WHAT IBM System Storage was, and WHY IBM should be considered when making a storage purchase decision. Let's recapsome of IBM's accomplishments during this time:
Today, October 1, I switch over to HOW to get it done. In my new job role, I will be leading a seriesof projects and workshops on how to make your data center more green, how to get more value from the information you have, how to better protect your information from unauthorized access or unethical tampering, how to develop and deploya site-wide business continuity plan, and how to centralize your management using open industry standards.
I will still be in Tucson, but am moving from building 9032 over to 9070 to be closer to the rest of my team.
Wrapping up this week's theme on why the System z10 EC mainframe can replace so many older, smaller,underutilized x86 boxes.This was all started to help fellow bloggers Jon Toigo of DrunkenData and Jeff Savit from Sun Microsystemsunderstand our IBM press release that we put out last February on this machine with my post[Yes, Jon, there is a mainframe that can help replace 1500 x86 servers] and my follow uppost [Virtualization, Carpools and Marathons"].The computations were based on running 1500 unique workloads as Linux guests under z/VM, and notrunning them as z/OS applications.
My colleagues in IBM Poughkeepsierecommended these books to provide more insight and in-depth understanding. Looks like some interesting summer reading. I put in quotes thesections I excerpted from the synopsis I found for each.
"From Microsoft to IBM, Compaq to Sun to DEC, virtually every large computer company now uses clustering as a key strategy for high-availability, high-performance computing. This book tells you why-and how. It cuts through the marketing hype and techno-religious wars surrounding parallel processing, delivering the practical information you need to purchase, market, plan or design servers and other high-performance computing systems.
Microsoft Cluster Services ("Wolfpack")
IBM Parallel Sysplex and SP systems
DEC OpenVMS Cluster and Memory Channel
Tandem ServerNet and Himalaya
Intel Virtual Interface Architecture
Symmetric Multiprocessors (SMPs) and NUMA systems"
Fellow IBM author Gregory Pfister worked in IBM Austin as a Senior Technical Staff Member focused on parallel processing issues, but I never met him in person. He points out that workloads fall into regions called parallel hell, parallel nirvana, and parallel purgatory. Careful examination of machine designs and benchmark definitions will show that the “industry standard benchmarks" fall largely in parallel nirvana and parallel purgatory. Large UNIX machines tend to be designed for these benchmarks and so are particularly well suited to parallel purgatory. Clusters of distributed systems do very well in parallel nirvana. The mainframe resides in parallel hell as do its primary workloads. The current confusion is where virtualization takes workloads, since there are no good benchmarks for it.
"In these days of shortened fiscal horizons and contracted time-to-market schedules, traditional approaches to capacity planning are often seen by management as tending to inflate their production schedules. Rather than giving up in the face of this kind of relentless pressure to get things done faster, Guerrilla Capacity Planning facilitates rapid forecasting of capacity requirements based on the opportunistic use of whatever performance data and tools are available in such a way that management insight is expanded but their schedules are not."
Neil Gunther points out that vendor claims of near linear scaling are not to be trusted and shows a method to “derate” scaling claims. His suggested scaling values for data base servers is closer IBM's LSPR-like scaling model, than TPC-C or SPEC scaling. I had mentioned that "While a 1-way z10 EC can handle 920 MIPS, the 64-way can only handle 30,657 MIPS."in my post, but still people felt I was using "linear scaling". Linear scaling would mean that if a 1Ghz single-core AMD Opteron can do four(4) MIPS, and an one-way z10 EC can do 920 MIPS, than one might assume that 1GHz dual-core AMD could do eight(8) MIPS, and the largest 64-way z10 EC can do theoretically 64 x 920 = 58,880 MIPS. The reality is closer to 6.866 and 30,657 MIPS, respectively.
This was never an IBM-vs-Sun debate. One could easily make the same argument that a large Sun or HP system could replace a bunch of small 2-way x86 servers from Dell. Both types of servers have their place and purpose, and IBMsells both to meet the different needs of our clients. The savings are in total cost of ownership, reducing powerand cooling costs, floorspace, software licenses, administration costs, and outages.
I hope we covered enough information so that Jeff can go back about talking about Sun products, and I can go backto talk about IBM storage products.