As I mentioned in my last post, [IBM announces the 60th anniversary of digital tape storage systems]. As promised, I am back in Tucson today to enjoy the festivities.
Ten years ago, I travelled to New York City with my colleague, Randy Fleenor, to present the latest in IBM tape technology for the 50th Anniversary. On Thursday evening that week, the latest movie in the Star Wars saga, Episode II: Attack of the Clones was just released, and it was being shown using the new Digital Light Projection (DLP) technology just around the corner at the Ziegfeld theater! This movie was the first live-action film to be filmed entirely digital. George Lucas saw that digital video was the future, and started the process moving forward with this film.
I convinced Randy to join me, and we arrived at 11:10pm, the movie was scheduled to start at 11pm, so we figured we had only missed a few previews. We walked into a completely empty lobby. I asked for two tickets for the 11pm show at the ticket counter, and was told it was all sold out, and there was a huge line around the building for all the people waiting to see the 1:00am show, and that we might get in to see the 3:00am show.
Randy and I had meetings on Friday morning, so we were not going to wait in line all night to see a 3am show! Just then, a young man comes out of the theater. He said his girlfriend can't make it, and wanted a refund for his two tickets. I pulled out a twenty-dollar bill, offered to buy them directly at face value, and the theater employees approved the transaction. The seats were front row of the balcony section. By then we had missed all the previews and a short bit of the movie, but that was alright with us.
(FTC Disclosure: I am both an employee and stockholder in IBM. The U.S. Federal Trade Commission may consider this a paid, celebrity endorsement of LTO-5 tapes and the LTFS technology. References to other companies are for illustrative purposes and do not represent an endorsement of their products or services.)
Digital recording is ideal for all types of video, including movies, television, and commercial advertisements.
The latest excitement is over IBM's Linear Tape File System™ (LTFS), which IBM donated to the IT industry as open source so that everyone in the world can benefit. This allows tape cartridges to be treated like USB memory sticks, the ultimate in portability of data. It is supported for Windows, Mac OS, and Linux, and already well embraced by the Media-and-Entertainment (M&E) industry.
Last week, IBM announced that it has helped [Greece's AlphaTV with LTFS on LTO-5 tape technology]. Here is an excerpt:
"The move to IBM technology has helped the network shrink its archive from 1,507 to just 388 square feet, representing dramatic systems and energy-cost savings."
"AlphaTV has been broadcasting since 1996, creating and storing all forms of video entertainment, from soap operas and documentaries, to movies and sporting events, and creating a vast video archive along the way. Initially, AlphaTV archived its programming on Sony Beta SP format video cassettes that stored up to 90 minutes of content. Not long after, in need of storage that offered greater density, it turned to DVCPRO format videos that stored up to 120 minutes. But even that format was not allowing the network to keep pace with its ballooning archive, a storage infrastructure that by 2011 spanned more than 1,507 square feet."
"'A Greek TV series stored on 100 DVCPRO tapes took up four shelves in our library, whereas on LTO-5 cartridge now takes up the space of a deck of playing cards,' Constantinos Colombus, chief technology officer at AlphaTV, said in a statement."
"IBM LTFS, an intuitive and graphical file system that provides direct access to data on LTO 5 drives, has enabled AlphaTV to manage, move and share video files much like they can with disk-management systems, by simply dragging and dropping. As a result, file management is easier to do and far more efficient, said Colombus."
To prepare for this anniversary, I spoke with Brad Johns, of [Brad Johns Consulting]. Brad was head of IBM tape marketing for a while, and ran tape customer councils to gather feedback from our largest customers. Brad was my mentor in marketing at IBM from 2003-2007 and has since retired from IBM to start his own consulting practice.
His latest publication is a TCO study entitled [A New Approach to Lowering the Cost of Storing File Archive Information] where he evaluated the 10-year total cost of ownership (TCO) for storage systems to hold file archives.
The comparison was made between Crossroad Systems' Strongbox® with Enterprise tape library, LTO-5 tapes using LTFS, versus a unified disk storage system offering NAS protocols on high-capacity 3TB drives. The findings: the tape-based archive had nearly 80 percent lower TCO than the disk-based solution!
You don't have to be in the middle of the Greek economy to real that is a good value!
technorati tags: IBM, DLP, Randy Fleenor, Star Wars, George Lucas, AlphaTV, LTO-5, tape, LTFS, TCO, Sony, DVCPRO
Well it's Tuesday again, and you know what that means.. IBM announcements! Today, IBM announces that next Monday marks the 60th anniversary of first commercial digital tape storage system! I am on the East coast this week visiting clients, but plan to be back in Tucson in time for the cake and fireworks next Monday.
Note that I said first commercial tape system as tape itself, in various forms, [has been around since 4000 B.C.]. A little historical context might help:
- 1845 - surgical tape
- 1921 - the Band-Aid, self-adhesive bandage
- 1925 - masking tape (which 3M sold under its newly announced Scotch® brand)
- 1930 - clear cellulose-based tape (today, when people say Scotch tape, they usually are referring to the cellulose version)
- 1935 - Allgemeine Elektrizitatsgesellschaft (AEG) presents Magnetophon K1, audio recording on analog tape
- 1942 - Duct tape
- 1947 - Bing Crosby adopts audio recording for his radio program. This eliminated him doing the same program live twice per day, perhaps the first example of using technology for "deduplication".
According to the IBM Archives the [IBM 726 tape drive was formally announced May 21, 1952]. It was the size of a refrigerator, and the tape reel was the size of a large pizza. The next time you pull a frozen pizza from your fridge, you can remember this month's celebration!
When I first joined IBM in 1986, there were three kinds of IBM tape. The round reel called 3420, and the square cartridge called 3480, and the tubes that contained a wide swath of tape stored in honeycomb shelves called the [IBM 3850 Mass Storage System].
My first job at IBM was to work on DFHSM, which was specifically started in 1977 to manage the IBM 3850, and later renamed to the DFSMShsm component of the DFSMS element of the z/OS operating system. This software was instrumental in keeping disk and tape at high 80-95 percent utilization rates on mainframe servers.
While visiting a client in Detroit, the client loved their StorageTek tape automation silo, but didn't care for the StorageTek drives inside were incompatible with IBM formats. They wanted to put IBM drives into the StorageTek silos. I agreed it was a good idea, and brought this back to the attention of development. In a contentious meeting with management and engineers, I presented this feedback from the client.
Everyone in the room said IBM couldn't do that. I asked "Why not?" The software engineers I spoke to already said they could support it. With StorageTek at the brink of Chapter 11 bankruptcy, I argued that IBM drives in their tape automation would ease the transition of our mainframe customers to an all-IBM environment.
Was the reason related to business/legal concerns, or was their a hardware issue? It turned out to be a little of both. On the business side, IBM had to agree to work with StorageTek on service and support to its mutual clients in mixed environments. On the technical side, the drive had to be tilted 12 degrees to line up with the robotic hand. A few years later, the IBM silo-compatible 3592 drive was commercially available.
Rather than put StorageTek completely out of business, it had the opposite effect. Now that IBM drives can be put in StorageTek libraries, everyone wanted one, basically bringing StorageTek back to life. This forced IBM to offer its own tape automation libraries.
In 1993, I filed my first patent. It was for the RECYCLE function in DFHSM to consolidate valid data from partial tapes to fresh new tapes. Before my patent, the RECYCLE function selected tapes alphabetically, by volume serial (VOLSER). My patent evaluated all tapes based on how full they were, and sorted them least-full to most-full, to maximize the return of cartridges.
Different tape cartridges can hold different amounts of data, especially with different formats on the same media type, with or without compression, so calculating the percentage full turned out to be a tricky algorithm that continues to be used in mainframe environments today.
The patent was popular for cross-licensing, and IBM has since filed additional patents for this invention in other countries to further increase its license revenue for intellectual property.
In 1997, IBM launched the IBM 3494 Virtual Tape Server (VTS), the first virtual tape storage device, blending disk and tape to optimal effect. This was based off the IBM 3850 Mass Storage Systems, which was the first virtual disk system, that used 3380 disk and tape to emulate the older 3350 disk systems.
In the VTS, tape volume images would be emulated as files on a disk system, then later moved to physical tape. We would call the disk the "Tape Volume Cache", and use caching algorithms to decide how long to keep data in cache, versus destage to tape. However, there were only a few tape drives, and sometimes when the VTS was busy, there were no tape drives available to destage the older images, and the cache would fill up.
I had already solved this problem in DFHSM, with a function called pre-migration. The idea was to pre-emptively copy data to tape, but leave it also on disk, so that when it needed to be destaged, all we had to do was delete the disk copy and activate the tape copy. We patented using this idea for the VTS, and it is still used in the successor models of IBM Sysem Storage TS7740 virtual tape libraries today.
Today, tape continues to be the least expensive storage medium, about 15 to 25 times less expensive, dollar-per-GB, than disk technologies. A dollar of today's LTO-5 tape can hold 22 days worth of MP3 music at 192 Kbps recording. A full TS1140 tape cartridge can hold 2 million copies of the book "War and Peace".
(If you have not read the book, Woody Allen took a speed reading course and read the entire novel in just 20 minutes. He summed up the novel in three words: "It involves Russia." By comparison, in the same 20 minutes, at 650MB/sec, the TS1140 drive can read this novel over and over 390,000 times.)
If you have your own "war stories" about tape, I would love to hear them, please consider posting a comment below.
technorati tags: IBM, AEG, Bing Crosby, Duct+Tape, Band Aid, DFHSM, RECYCLE, DFSMShsm, z/OS, StorageTek, VTS, VTL, LTO-5, TS1140, LTFS, Woody Allen
IBM wins lots of awards, but this time is unique: [IBM and Fox Networks Group have jointly won an Engineering Emmy® Award] for Innovation from the Academy of Television Arts & Sciences. According to the Academy, by improving the ability of media companies to capture, manage and exploit content in digital form, IBM and Fox have fundamentally changed the way that audio and video content is managed and stored. Here's an excerpt from the IBM Press Release:
"By standardizing technologies in this way, Fox can now use open-standard, file-based tape in all aspects of production, post-production and distribution functions – displacing costly proprietary tape formats and/or disk subsystems. This provides media companies with the consumer equivalent of having their entire library of DVDs online and available at any time, and the ability to go to a specific scene, in any one of the movies, in an instant.
In the early stage of the technology initiative, the IBM/Fox team applied IBM-patented technologies invented by IBM Research for high-speed data movement. They also integrated traditional broadcast transport and encoding standards with IT industry open standards. This allowed either Standard Definition (SD) or HD programming to be available in real time for digital recording and repurposing -- with improved economics."
Last year, in my posts [Double the Storage Capacity - Double the Fun!], [Happy E.A.R.T.H day!], [Local IBM Team recognized by Arizona Daily Star], and [On Cirago Docking Stations, IBM RDX and LTFS], I had referred to LTFS by its prior name, the Long Term File System. People keep archives on tape because it is a great medium for long term retention.
Unfortunately, people didn't like the name, but they loved the acronym, so it was renamed to Linear Tape File System. IBM offers LTFS single-drive edition on its LTO-5 and TS1140 tape drives, and LTFS library-edition across all of its tape libraries. Since everyone hates proprietary vendor lock-in, IBM has graciously shared LTFS as an open source standard with the rest of the Linear Tape Open consortium.
(Note: I was not there at the awards ceremony. The pictures were taken by Ed Childers, David Pease and Rainer Richter of each other. Additional photos are available on this [Flicr photo album].)
(1) Rainer Richter, Media Technology Market Partners LLC [MTMP], presenting the Emmy to Steve Canepa, IBM General Manager for Media and Entertainment industry. MTMP is an IBM Business Partner that offers integrated solutions for LTO and LTFS, consulting, services, and technology to the media and entertainment industry…
(2) Ed Childers, IBM manager of the Tape Drive Development team, holding the Emmy. Fellow IBM blogger Steve Hamm credits Ed on coming up with the idea for LTFS seven years ago, in his blog post [Coding and Loading in Las Vegas: How a Team of IBM Researchers Helped Transform the Way Video is Stored]. Ed wanted to make tape storage easier to use and to integrate it into the workflow of networks and studios, and suggested using an indexing system that would allow people to write software that would make video more accessible.
(3) David Pease, IBM Senior Technical Staff Member from the IBM Almaden Research Center, holding the Emmy. Along with Lucas Villa Real (IBM Brazil) and Michael Richmond (IBM Almaden), David and his team were able to come up with a working prototype in just four months. Michael discusses this in his posts [Tape? Does anyone care about Tape anymore?"] and [the Emmy goes to... LTFS].
If David and Ed look familiar, it is because I had their picture on my blog back in May when [LTFS won 2011 NAB Show Pick Hits award]. For more from the IBM Almaden Research team, see their blog post [We won an Emmy!]
Of course, Technology is only worthwhile if you put it to use. Our friends at FOX initially partnered with IBM to develop this video archive solution for the National Football League (NFL). If there is one place that "re-purposes" a lot of video footage, it is sports television. The technology proved so useful that FOX has since expanded it to other types of programming.
technorati tags: IBM, FOX, Emmy, Ed Childers, David Pease, Steve Canepa, Michael Richmond, Rainer Richter, MTMP, Almaden Research Center, LTFS, LTO-5, TS1140, Media and Entertainment, Steve Hamm
This year marks the 10 year anniversary of IBM's introduction of LTO tape technology. IBM is a member of the Linear Tape Open consortium which consists of IBM, HP and Quantum, referred to as "Technology Provider Companies" or TPCs. In an earlier job role, I was the "portfolio manager" for both LTO and Enterprise tape product lines.
Today, we held a celebration in Tucson, with cake and refreshments.
IBM Executives Doug Balog, IBM VP of Storage Platform, and Sanjay Tripathi, the new IBM Director and Business Line Executive for Tape, VTL and Archive systems, presented the successes of LTO tape over the past 10 years.
To date over 3.5 million LTO tape drives, and over 150 million LTO tape media cartridges have been shipped which is a testament to the remarkable marketplace acceptance of the technology.
In honor of this event, I decided to interview Bruce Master, IBM Senior Program Manager for Data Protection Systems, about this 10 year anniversary.
10 years of LTO technology is a great milestone. How is this especially significant to IBM and its clients?
According to IDC data, IBM has held the #1 leader position in market share for total world wide branded tape revenue for over 7 years and that IBM is still #1 in branded midrange tape revenue which includes the LTO tape technologies. IBM was the first drive manufacturer to deliver LTO-1 drives, back in September 2000, the first to deliver tape drive encryption to the marketplace on LTO-4 drives, and is shipping LTO generation 5 drives and libraries. IBM is the author of the new Linear Tape File System (LTFS) specification that has been adopted by the TPCs. This file system revolutionizes how tape can be used as if it were a giant 1.5 terabyte removable USB memory stick with the capability to be accessed with directory tree structures and drag and drop functionality. With LTO's built-in real-time compression, a single tape cartridge can hold up to 3TB of data.
The Linear Tape File System has been getting a lot of attention. Where can we learn more about it?
Researchers at IBM's Almaden Research Center invented the [Linear Tape File System], released it as Open Source under the name [IBM Long Term File System], and contributed the specification to the LTO consortium. On the [Ultrium.com] website, you can read articles written about the file system, the specification [60-page PDF] document and a [video demo] of the file system in action. There is also an article out on [Wikipedia].
Why is tape still a critical part of a storage infrastructure?
Tape is low cost and provides critical off-line portable storage to help protect data from attacks that can occur with on-line data. For instance, on-line data is at risk of attack from a virus, hacker, system error, disgruntled employee, and more. Since tape is off-line, not accessible by the system, it protects against these forms of corruption. LTO technology also provides write-once read-many (WORM) tape media to help address compliance issues that specify non-erasable, non-rewriteable (NENR) storage, hardware encryption to secure data, as well as a low cost long term archive media. When data cools off, or becomes infrequently accessed, why keep it on spinning disk? Move it to tape where it is much greener and lower cost. A tape in a slot on a shelf consumes minimal energy.
So tape is not dead?
Ha! Far from it. Seems like disk-only "specialty shop" storage vendors that don’t have tape in their sales portfolio are the ones that propagate that myth. In reality, storage managers are tasked with meeting complex objectives for performance, compliance, security, data protection, archive and total cost of ownership. Optimally, a blend of disk and tape in a tiered infrastructure can best address these objectives. You can’t build a house with just a hammer. IBM has a rich tool kit of storage offerings including disk, tape, software, services and deduplication technologies to help clients address their needs.
Do you have an example of a client who was saved by tape?
Yes indeed. Estes Express, a large trucking firm, was hit by a hurricane that flooded their data center and destroyed all systems. Fortunately the company survived because the night before they had backed up all data on to IBM tape and moved the cartridges offsite! The company survived and has since implemented a best practices data protection strategy with a combination of disk-to-disk-to-tape (D2D2T) using LTO tape at the primary site, and a remote global mirrored site that is also backed up to LTO tape.
So tape saved the day. What is the outlook for tape innovation in the future?
The future is bright for tape. Earlier this year, IBM and Fujifilm were able to [demonstrate a tape density achievement] that could enable a native 35TB tape cartridge capacity! This shows a long roadmap ahead for tape and a continued good night’s sleep for storage managers knowing that their precious data will be safe.
Of course, LTO tape is just one of the many reasons IBM is a successful and profitable leader in the IT storage industry. Doug Balog talked about his experiences in London for the [October 7th launch] of IBM DS8800, Storwize V7000 and SAN Volume Controller 6.1. Sanjay Tripathi showed recent successes with IBM's ProtecTIER Data Deduplication Solution and Information Archive products.
I would like to thank Bruce Master for his time in completing this interview. To learn more about IBM tape and storage offerings, visit [ibm.com/storage].
technorati tags: IBM, Linear Tape Open, LTO, LTO-1, LTO-2, LTO-3, LTO-4, LTO-5, Doug Balog, Sanjay Tripathi, Bruce Master
In my presentations in Australia and New Zealand, I mentioned that people were re-discovering the benefits of removable media. While floppy diskettes were convenient way of passing information from one person to another, they unfortunately did not have enough capacity. In today's world, you may need Gigabytes or Terabytes of re-writeable storage with a file system interface that can easily be passed from one person to another. In this post, I explore three options.
- Cirago CDD2000 Docking Station
The good folks over at [Cirago International Ltd.] sent me a cute little [CDD2000 docking station] for evalution.
(FCC Disclaimer: I work for IBM, and IBM has no business relationship with Cirago at the time of this writing. Cirago has not paid me to mention their product, but instead provided me a free loaner that I promised to return to them after my evaluation is completed. This post should not be considered an endorsement for Cirago's products. List prices for Cirago and IBM products were determined from publicly available sources for the United States, and may vary in different countries. The views expressed herein may not necessarily reflect the views and opinions of either IBM or Cirago.)
I took a few photos so you can see what exactly this device looks like. Basically, it is a plastic box that holds a single naked disk drive. It has four little rubber feet so that it does not slip on your desk surface.
The inside is quite simple. The power and SATA connections match those of either a standard 3.5 inch drive, or the smaller form factor (SFF) 2.5 inch drive. However, to my dismay, it does not handle EIDE drives which I have a ton of. After taking apart six different computer systems, I found only one had SATA drives for me to try this unit out with.
The unit comes with a USB cable and AC/DC power adapter. In my case, I found the USB 3.0 cable too short for my liking. My tower systems are under my desk, but I like keeping docking stations like this on the top of the desk, within easy reach, but that wasn't going to happen because the USB cable was not long enough.
Instead, I ended up putting it half-way in between, behind my desk, sitting on another spare system. Not ideal, but in theory there are USB-extension cables that probably could fix this.
Here it is with the drive inside. I had a 3.5 inch Western Digital [1600AAJS drive] 160 GB, SATA 3 Gbps, 8 MB Cache, 7200 RPM.
To compare the performance, I used a dual-core AMD [Athlon X2] system that I had built for my 2008 [One Laptop Per Child] project. To compare the performance, I ran with the drive externally in the Cirago docking station, then ran the same tests with the same drive internally on the native SATA controller. Although the Cirago documentation indicated that Windows was required, I used Ubuntu Linux 10.04 LTS just fine, using the flexible I/O [fio] benchmarking tool against an ext3 file system.
- Sequential Write - a common use for external disk drive is backup.
- Random read - randomly read files ranging from 5KB to 10MB in size.
- Random mixed - randomly read/write files (50/50 mix) ranging from 5KB to 10MB in size.
|Sequential Write||Throughput IOPS||1119||1044|
| ||Latency (msec)||0.866 ms||0.948 ms|
| ||Bandwidth (KB/s)||16900||14400|
|Random Read||Throughput (IOPS)||164||119|
| ||Latency (msec)||6.06 ms||8.36 ms|
| ||Bandwidth (KB/s)||658||477|
|Random Mixed (50/50)||Throughput (IOPS)||112||81|
| ||Latency (msec) read||8.78 ms||12.1 ms|
| ||Latency (msec) write||0.0983 ms||0.120 ms|
| ||Bandwidth (KB/s) read||557||328|
| ||Bandwidth (KB/s) write||556||337|
For sequential write, the Cirago performed well, only about 15 percent slower than native SATA. For random workloads, however, it was 30-40 percent slower. If you are wondering why I did not get USB 3.0 speeds, there are several factors involved here. First, with overheads, 5 Gbps USB 3.0 is expected to get only about 400 MB/sec. My SATA 2.0 controller maxes out at 375 MB/sec, and my USB 2.0 ports on my system are rated for 57 MB/sec, but with overheads will only get 20-25 MB/sec. Most spinning drives only get 75 to 110 MB/sec. Even solid-state drives top out at 250 MB/sec for sustained activity. Despite all that, my internal SATA drive only got 16 MB/sec, and externally with the Cirago 14 MB/sec in sustained write activity.
Here is the mess that is inside my system. The slot for drive 2 was blocked by cables, memory chips and the heat sink for my processor. It is possible to damage a system just trying to squeeze between these obstacles.
However, the point of this post is "removable media". Having to open up the case and insert the second drive and wire it up to the correct SATA port was a pain, and certainly a more difficult challenge than the average PC user wishes to tackle.
Price-wise, the Cirago lists for $49 USD, and the 160GB drive I used lists for $69, so the combination $118 is about what you would pay for a fully integrated external USB drive. However, if you had lots of loose drives, then this could be more convenient and start to save you some money.
- IBM RDX disk backup system
Another problem with the Cirago approach is that the disk drives are naked, with printed circuit board (PCB) exposed. When not in the docking station, where do you put your drive? Did you keep the [anti-static ESD bag] that it came in when you bought it? And once inside the bag, now what? Do you want to just stack it up in a pile with your other pieces of equipment?
To solve this, IBM offers the RDX backup system. These are fully compatible with other RDX sytems from Dell, HP, Imation, NEC, Quantum, and Tandberg Data. The concept is to have a docking station that takes removable, rugged plastic-coated disk-enclosed cartridges. The docking station can be part of the PC itself, similar to how CD/DVD drives are installed, or as a stand-alone USB 2.0 system, capable of processing data up to 25 MB/sec.
The idea is not new, about 10 years ago we had [Iomega "zip" drives] that offered disk-enclosed cartridges with capacities of 100, 250 and 750MB in size. Iomega had its fair share of problems with the zip drive, which were ranked in 2006 as the 15th worst technology product of all time, and were eventually were bought out by EMC two years later (as if EMC has not had enough failures on its own!)
The problem with zip drives was that they did not hold as much as CD or DVD media, and were more expensive. By comparison, IBM RDX cartridges come in 160GB to 750GB in size, at list prices starting at $127 USD.
- IBM LTO tape with Long-Term File System
Removable media is not just for backup. Disk cartridges, like the IBM RDX above, had the advantage of being random access, but most tape are accessed sequentially. IBM has solved this also, with the new IBM Long Term File System [LTFS], available for LTO-5 tape cartridges.
With LFTS, the LTO-5 tape cartridge now can act as a super-large USB memory stick for passing information from one person to the next. The LTO-5 cartridge can handle up to 3TB of compressed data at up to SAS speeds of 140 MB/sec. An LTO-5 tape cartridge lists for only $87 USD.
The LTO-5 drives, such as the IBM [TS2250 drive] can read LTO-3, LTO-4 and LTO-5cartridges, and can write LTO-4 and LTO-5 cartridges, in a manner that is fully compatible with LTO drives from HP or Quantum. LTO-3, LTO-4 and LTO-5 cartridges are available in WORM or rewriteable formats. LTO-4 and LTO-5 cartridges can be encrypted with 256-bit AES built-in encryption. With three drive manufacturers, and seven cartridge manufacturers, there is no threat of vendor lock-in with this approach.
These three options offer various trade-offs in price, performance, security and convenience. Not surprisingly, tape continues to be the cheapest option.
technorati tags: IBM, Cirago, CDD2000, RDX, Ubuntu, Linux, LTO, LTO-5, LTFS, SATA, USB, fio