In a previous post I mentioned that I’ve been talking to TV and film production houses for an upcoming supplement called Storage in the Studio. Often, these entertainment facilities solve their unique storage challenges (the need for blazing bandwidth and capacious capacity) with “me-too” “white-box” disk subsystems that use conventional interconnects such as Fibre Channel or Ethernet.
But Advanced Digital Services, a Hollywood-based post-production and media services studio, opted for a horse of a different color. The company is using JMR’s PeSAN architecture, which I think (correct me if I’m wrong) is unique in that it uses the ubiquitous, high-speed PCIe bus to connect hosts to JMR’s BlueStor disk arrays. Using the PCIe bus as a storage interconnect eliminates the overhead associated with Fibre Channel, Infiniband or Ethernet protocol conversion.
Russell Ruggieri, ADS’ director of engineering, says that the BlueStor-PeSAN combo is less expensive that traditional alternatives (JMR cites pricing of less than $1 per GB) and provides plenty of bandwidth for the studio’s non-linear editing work.
Ruggieri says the new storage configuration solved the performance problems associated with his Apple Xsan configuration, providing more than 1,000MBps of throughput and the ability to handle 4K-resolution formats. (JMR claims performance of more than 2,000MBps.) “It also makes it easy to add capacity, and JMR’s PCI adapters give me additional slots on the Macs,” says Ruggieri. ADS may replace the remainder of its 2Gbps Fibre Channel configurations with additional BlueStor systems in a PeSAN configuration.
I had heard about the PeSAN technology about four years ago when it was in development, but I was skeptical. (If it was such a good idea, why hadn’t anybody else thought of it?) But JMR may be on to something with this architecture. The company seems to be gaining momentum in application areas such as content creation, SD/HD video editing, and 2K/4K digital intermediate applications.
Tuesday, February 24, 2009
Thursday, February 19, 2009
Seagate drops STEC lawsuit
You know you’ve hit the big time when you get sued by Seagate. That’s what happened to solid-state disk (SSD) specialist STEC last April when Seagate filed a patent infringement lawsuit against the company.
Today, that lawsuit was dropped. In legalese, it was a “mutual dismissal,” whatever that implies. No cash exchanged hands, and neither company licensed its technology to the other.
Not that many people thought that Seagate would prevail in this lawsuit (which had something to do with misappropriation of intellectual property relating to hard disk technology), but it does give the green light to STEC. This relatively small company from my neck of the woods (southern California) seems to have a lot of momentum (at least to the degree that enterprise-class SSDs have a lot momentum), having inked OEM deals with the likes of EMC, Sun and, most recently, Hitachi Data Systems (and more to come in the very near future) for its ZeusIOPS SSDs.
Related articles from infostor.com
HDS puts SSDs in disk arrays
Sun ships arrays with SSDs
Who uses SSDs, and why?
SNIA launches SSD initiative
SSDs poised for the enterprise
Today, that lawsuit was dropped. In legalese, it was a “mutual dismissal,” whatever that implies. No cash exchanged hands, and neither company licensed its technology to the other.
Not that many people thought that Seagate would prevail in this lawsuit (which had something to do with misappropriation of intellectual property relating to hard disk technology), but it does give the green light to STEC. This relatively small company from my neck of the woods (southern California) seems to have a lot of momentum (at least to the degree that enterprise-class SSDs have a lot momentum), having inked OEM deals with the likes of EMC, Sun and, most recently, Hitachi Data Systems (and more to come in the very near future) for its ZeusIOPS SSDs.
Related articles from infostor.com
HDS puts SSDs in disk arrays
Sun ships arrays with SSDs
Who uses SSDs, and why?
SNIA launches SSD initiative
SSDs poised for the enterprise
Tuesday, February 17, 2009
IT energy efficiency -- the book
I just received the latest book (yes, an actual printed book) written by my buddy Greg Schulz. (No, this isn’t a plug for a book from my buddy – Greg is a well-known storage industry analyst and founder of storageio.com.)
Although Greg is known primarily for his storage knowledge, The Green and Virtual Data Center (from CRC Press) covers energy efficiency and virtualization across all areas of IT. This is essential reading for any IT manager that is serious about “going green.” (According to Greg’s research, only about 5% to 10% of IT organizations have a green initiative, need or mandate, although 75% to 85% cite problems such as reliable availability of electrical power and limits on cooling capacities and floor space.)
Some of the broader technologies covered in the 376-page book include energy and cost footprint reduction, cloud-based storage and computing, managed services, intelligent power management and adaptive power management, blade centers and servers, virtualization and tiering (of servers, storage and networks), data reduction, compression, data de-duplication, energy avoidance, and many others, including storage-specific technologies such as MAID (massive array of idle disks) and thin provisioning.
Check it out here.
Although Greg is known primarily for his storage knowledge, The Green and Virtual Data Center (from CRC Press) covers energy efficiency and virtualization across all areas of IT. This is essential reading for any IT manager that is serious about “going green.” (According to Greg’s research, only about 5% to 10% of IT organizations have a green initiative, need or mandate, although 75% to 85% cite problems such as reliable availability of electrical power and limits on cooling capacities and floor space.)
Some of the broader technologies covered in the 376-page book include energy and cost footprint reduction, cloud-based storage and computing, managed services, intelligent power management and adaptive power management, blade centers and servers, virtualization and tiering (of servers, storage and networks), data reduction, compression, data de-duplication, energy avoidance, and many others, including storage-specific technologies such as MAID (massive array of idle disks) and thin provisioning.
Check it out here.
Saturday, February 7, 2009
200TB and 413,138 hairs
I’ve been talking to film and TV pre- and post-production outfits for an upcoming supplement we call Storage in the Studio, which includes case studies of studios that have solved somewhat unique storage problems. I try to focus on the storage angle, but a lot of interesting stuff invariably comes up.
For example, in creating graphics for the feature film The Tale of Despereaux, UK-based Framestore started with 50TB of capacity and by the end of production the studio had a whopping 200TB -- 400TB if you include a mirrored cluster. (All of the content was stored on Infortrend’s EonStor disk arrays in a, somewhat surprisingly, RAID-6 configuration for added reliability.). During peak production, Framestore was generating 5TB of animation content per day.
The computer generated imagery (CGI) studio created the 200TB storage cluster using the Lustre open-source Linux cluster file system. The cluster was tightly coupled to a 6,000-core render farm. Fun facts:
--There are 40,089 individual assets in the film
--1,726 final shots were delivered
--The film has 126,248 frames
--The animation crew size was 278 during peak production
And there are 413,138 hairs on Despereaux’s head.
For example, in creating graphics for the feature film The Tale of Despereaux, UK-based Framestore started with 50TB of capacity and by the end of production the studio had a whopping 200TB -- 400TB if you include a mirrored cluster. (All of the content was stored on Infortrend’s EonStor disk arrays in a, somewhat surprisingly, RAID-6 configuration for added reliability.). During peak production, Framestore was generating 5TB of animation content per day.
The computer generated imagery (CGI) studio created the 200TB storage cluster using the Lustre open-source Linux cluster file system. The cluster was tightly coupled to a 6,000-core render farm. Fun facts:
--There are 40,089 individual assets in the film
--1,726 final shots were delivered
--The film has 126,248 frames
--The animation crew size was 278 during peak production
And there are 413,138 hairs on Despereaux’s head.
Tuesday, February 3, 2009
Consider data de-duplication plus compression
Data de-duplication gets a lot of ink and engenders a lot of debates, but the option of combining data de-dupe with real-time compression on primary storage is rarely discussed, maybe because there aren’t many vendors targeting compression for primary storage resources.
That’s what Storwize specializes in, and the company recently released the results of tests that suggest a serious look at combining de-duplication with data compression to further enhance overall data reduction.
Storwize claims that adding real-time compression can improve data reduction by more than 200%, based on its internal testing as well as customer deployments. (In the internal tests, Storwize used data de-duplication solutions from both Data Domain and NetApp.)
Storwize positions data de-duplication as being particularly advantageous for highly redundant backup data sets, but points out that using real-time data compression on the front end (primary storage devices) can significantly reduce de-duplication processing time and enhance throughput, in addition to reducing overall capacity requirements.
To assess the validity of the test results, check them out here
And if you’re among the early adopters of data compression on primary storage, respond below or email me at daves@pennwell.com and let me know what types of real-life data compression ratios you’re getting.
That’s what Storwize specializes in, and the company recently released the results of tests that suggest a serious look at combining de-duplication with data compression to further enhance overall data reduction.
Storwize claims that adding real-time compression can improve data reduction by more than 200%, based on its internal testing as well as customer deployments. (In the internal tests, Storwize used data de-duplication solutions from both Data Domain and NetApp.)
Storwize positions data de-duplication as being particularly advantageous for highly redundant backup data sets, but points out that using real-time data compression on the front end (primary storage devices) can significantly reduce de-duplication processing time and enhance throughput, in addition to reducing overall capacity requirements.
To assess the validity of the test results, check them out here
And if you’re among the early adopters of data compression on primary storage, respond below or email me at daves@pennwell.com and let me know what types of real-life data compression ratios you’re getting.
Friday, January 30, 2009
EMC racks up record results
You gotta hand it to the folks in Hopkinton: In what might turn out to be the worst IT spending slump since 2001—2002, EMC managed to post record-setting financial results not only for 2008 but also for the fourth quarter.
This week, the company announced that its 2008 revenues hit $14.88 billion – up 12% over 2007. Q4 2008 revenues topped $4 billion – up 8% sequentially and the first time EMC topped $4 billion in a quarter (not to mention generating an operating cash flow of $1.1 billion and free cash flow of $775 million). And it was the company’s sixth consecutive year of double-digit revenue growth.
On the storage front, the company reported that its Information Infrastructure business (which includes storage, RSA security, and content management and archiving products) racked up $3.5 billion in the fourth quarter – an increase of 8% sequentially. (VMware contributed revenue of $514 million in the fourth quarter, and $188 billion for the year.)
Storage-specific revenue increased 4% in 2008, in spite of weak Symmetrix performance (down 9%). That compares to IBM’s 16% decline in storage revenues in 2008, according to Robert W. Baird & Co.
But beyond the stellar financial results (given the macro economic conditions), I’m even more impressed with the company’s cost cutting measures, which recently included layoffs of about 2,400 employees, or approximately 7% of the company’s workforce, and a restructuring that included consolidating back-office functions, reduced spending on contractors, and a reduction in “management layers.” EMC officials expect the restructuring to reduce the company’s cost base by $350 million this year.
During the announcement, EMC officials estimated that 2009 global spending will decline as a percentage in the mid to high single digits compared to 2008, and that a higher-than-usual percentage of IT spending will take place in the second half of the year.
This week, the company announced that its 2008 revenues hit $14.88 billion – up 12% over 2007. Q4 2008 revenues topped $4 billion – up 8% sequentially and the first time EMC topped $4 billion in a quarter (not to mention generating an operating cash flow of $1.1 billion and free cash flow of $775 million). And it was the company’s sixth consecutive year of double-digit revenue growth.
On the storage front, the company reported that its Information Infrastructure business (which includes storage, RSA security, and content management and archiving products) racked up $3.5 billion in the fourth quarter – an increase of 8% sequentially. (VMware contributed revenue of $514 million in the fourth quarter, and $188 billion for the year.)
Storage-specific revenue increased 4% in 2008, in spite of weak Symmetrix performance (down 9%). That compares to IBM’s 16% decline in storage revenues in 2008, according to Robert W. Baird & Co.
But beyond the stellar financial results (given the macro economic conditions), I’m even more impressed with the company’s cost cutting measures, which recently included layoffs of about 2,400 employees, or approximately 7% of the company’s workforce, and a restructuring that included consolidating back-office functions, reduced spending on contractors, and a reduction in “management layers.” EMC officials expect the restructuring to reduce the company’s cost base by $350 million this year.
During the announcement, EMC officials estimated that 2009 global spending will decline as a percentage in the mid to high single digits compared to 2008, and that a higher-than-usual percentage of IT spending will take place in the second half of the year.
Friday, January 23, 2009
Application-centric storage challenges SANs
I’m in the process of editing an interesting, and perhaps controversial, article by Andrew Reichman, a senior analyst at Forrester Research. It’s based on his report, “Do You Really Need a SAN Anymore?” and makes the argument that application-centric storage may prove to be a viable alternative to traditional SAN architectures.
“Application-centric storage” is when application vendors embed storage functionality into their code to handle some, or many, of the functions typically handled by disk array vendors or third-party software vendors. As such, storage is managed from within the application itself.
I sincerely doubt that application-centric storage will challenge SANs, but in some cases it could make a lot of sense. And don’t underestimate the storage plans or capabilities of vendors such as VMware, Microsoft and Oracle.
Here are some examples, which I’m lifting from the article:
Oracle offers branded storage hardware under the Exadata label. Using its Automatic Storage Management (ASM) feature to control the flow of data between application servers, and storage servers provided by technology partner HP, the Exadata system is a hybrid of software and industry standard hardware that Oracle will sell directly.
Microsoft recommends DAS for Exchange 2007. Exchange has long had a best practice for configuration that precluded adding other workloads to the storage array it runs on, which has served to put it on an island. Now, with the addition of Cluster Continuous Replication (CCR), the application can manage its own high availability and replication.
VMware offers native storage management capabilities. The virtualization market leader continues to add storage features to its application stack. For example, VMFS offers volume management capabilities within virtual machine management, and VMware recently announced vStorage, which includes native thin provisioning and other key storage features. While many of the storage features within VMware leverage SAN array capabilities, the control and management is substantially shifted to the application realm.
Leading application vendors probably won’t rock the SAN boat, but it’s quite possible that they’ll provide some worthy competition to the pure-play storage hardware/software vendors.
“Application-centric storage” is when application vendors embed storage functionality into their code to handle some, or many, of the functions typically handled by disk array vendors or third-party software vendors. As such, storage is managed from within the application itself.
I sincerely doubt that application-centric storage will challenge SANs, but in some cases it could make a lot of sense. And don’t underestimate the storage plans or capabilities of vendors such as VMware, Microsoft and Oracle.
Here are some examples, which I’m lifting from the article:
Oracle offers branded storage hardware under the Exadata label. Using its Automatic Storage Management (ASM) feature to control the flow of data between application servers, and storage servers provided by technology partner HP, the Exadata system is a hybrid of software and industry standard hardware that Oracle will sell directly.
Microsoft recommends DAS for Exchange 2007. Exchange has long had a best practice for configuration that precluded adding other workloads to the storage array it runs on, which has served to put it on an island. Now, with the addition of Cluster Continuous Replication (CCR), the application can manage its own high availability and replication.
VMware offers native storage management capabilities. The virtualization market leader continues to add storage features to its application stack. For example, VMFS offers volume management capabilities within virtual machine management, and VMware recently announced vStorage, which includes native thin provisioning and other key storage features. While many of the storage features within VMware leverage SAN array capabilities, the control and management is substantially shifted to the application realm.
Leading application vendors probably won’t rock the SAN boat, but it’s quite possible that they’ll provide some worthy competition to the pure-play storage hardware/software vendors.
Subscribe to:
Posts (Atom)