Cleversafe receives Wikibon CTO award for the Best Storage Technology Innovations of 2009

Cleversafe has been recognized as a leader by Wikibon industry veterans with the Wikibon CTO award for the Best Storage Technology Innovations of 2009. Here’s what Wikibon had to say about Cleversafe:

Wikibon believes that this type of technology is a good strategic fit for cloud storage, especially for archive and media data. The overhead for splitting it up in to 16 slices is 25%, making it half the price of a traditional protected storage. The alternative is no backup at all for cloud storage, an approach being more popular but harder to detect.

Cleversafe introduced dispersed storage technology commercially in 2008 after three years of research and development. This award validates the enhancements Cleversafe has made since product launch. Products that meet cloud requirements are not going to be built overnight, and Cleversafe has demonstrated diligent and progressing efforts to serve the requirements that a distributed, multi-tenancy storage platform demand.

We are also pleased with our product portfolio progress. Look for announcements in the next month from us that further validate Cleversafe as a leader in massively scalable systems designed with inherent data security, particularly for data in the cloud.

Click here for the full post at Wikibon.

Other winners include Storwize, and Unisys.

Thoughts on Mexico Federal Government Conference

On January 20th, I attended a conference on cloud computing and cloud storage hosted by the Mexican federal government who is looking at how they can use cloud systems to provide more efficient and more reliable IT systems both for internal use as well as to better deliver services to its citizens.  They selected eight leading cloud computing and cloud storage vendors to present their cloud strategies:  IBM, EMC, Google, Sun, HP, Cisco, Hitachi, and Cleversafe.  There’s been a lot of press coverage lately of “cloud”, so it was quite interesting to really see the presentations to directly understand the state of cloud technology and what each company is providing.

Surprisingly, six of these companies were consistent in not discussing how their products have evolved to actually address the requirements of the cloud.

Was it designed to work in the cloud?

IBM presented a definition of cloud computing and spoke very generally about the elements and benefits of cloud computing.  Although IBM didn’t present any new technologies for cloud, they have a very long history in the technology approaches leading up to cloud and are integrating various solutions for the cloud market.

Carlos Medinas of Cisco presented a unified architecture for cloud computing which promised higher efficiency.  Understandably, Cisco views the world as a series of detailed network diagrams which results in a presentation of a series of detailed network diagrams.  Clearly, cloud computing and cloud storage will utilize the Internet network infrastructure provided primarily by Cisco, so they will play a key role; however, Cisco will need to work with various partners to delivery solution-level benefits.

Antonio Guerraro of HP (who was formerly in EDS Mexico) then spoke on the business process capabilities of cloud computing.  HP’s three strategies are to:

-          Help customers secure, manage and administer cloud systems

-          Enable providers of cloud services

-          Provide cloud services through partners like Gobble, Print 2 Cloud, MacCloud, Snapfish, etc.

At this point in the presentation, it became clear that the standard cloud presentation was to 1) define the domain of cloud computing, 2) talk about how your company has been working on cloud-like capabilities since the dawn of (digital) time and then 3) show a big matrix of capabilities that shows that your company knows everything and can do everything – thus providing a complete solution.  As had IBM and Cisco, HP presented this approach which then was repeated by Sun, EMC and Hitachi.

Sun’s unique spin was that cloud-like capabilities was so integral to the company was that it was their company slogan, “The Network is the Computer”.  No doubt Sun has always been on this theme and has played a leadership role with technologies like Java, Ethernet workstations, etc.  With the Oracle merger concluding, it will be interesting to see how that affect’s Sun’s (and Oracle’s) strategy.

EMC understandably featured their virtual computing capabilities through VMware as well as their partnership with Cisco and VMware for providing cloud solutions.  They then positioned their Atmos product as COS – Cloud Optimized Storage – a new a category of storage like NAS or SAN.  EMC touted the capability of Atmos to manage data sets using policies.  For example, you can designate policy levels of “bronze” and “gold” which would each correspond to different reliability levels with higher numbers of copies for data associated with the gold policy.  EMC said that Atmos would automatically make multiple copies as needed which they illustrated by showing a map of Mexico with “4, 5 or ‘n’ copies” scattered across the map.  I can certainly see why EMC would want to sell a system that automatically makes 4, 5 or ‘n’ copies, but it is hard to see how most customers would find that approach cost effective.

As with all the prior presenters, Hitachi Data Systems lead off with defining the cloud storage market.  Hitachi’s definition included three segments:  public cloud, private cloud and content repository systems.  Hitachi asserted their differentiator as being able to provide different storage interfaces – NAS and SAN – on the same pool of drives.  Hitachi also has a “dynamic provisioning” system and retention (compliance) policies and position themselves around the differentiators of integration, reliability & security and scalability.

Each of these prior six presenters told a similar story and it seems as though many vendors are talking about how they have the breadth and depth to cover cloud, but aren’t giving any concrete examples of how their products have fundamentally evolved to address the requirements of a massively scalable, multi-tenancy system.

John Farrell of Google was up next and it was refreshing that he had something different to say.  He started at a very high level by referring to the book The Big Switch by Nicholas Carr.  The general theme was that computing and applications are much more efficient when delivered as a utility, like electricity.  John also provided some insights into the improved cost efficiency – specifically the life cycle cost – of gmail vs. operating an internal email system.

I then concluded the conference by stepping back and looking at long term technology trends to answer the question of whether cloud computing and cloud storage is just an industry fad or a genuine shift.  My presentation then used this technology megatrends framework to look in detail at where cloud storage is headed and how we’ll get there which is the subject a future post…

Cleversafe Ranks Third in StorageMonkeys Top Storage Vendor Blogs – 2010

Cleversafe is pleased to announce our 3rd place finish in the StorageMonkeys Top Storage Vendors Blogs for 2010. You may be confused based on the official results posted on StorageMonkeys that placed Cleversafe in 21st.

We are not calling for a revote, but perhaps a more equitable calculation is in order, you review the math…

Rank Company Blog Votes Employee #* %
1 Zetta Zetta blog 39 25 156.00%
2 Nirvanix Stephen Foskett 55 38 144.74%
3 Cleversafe Cleversafe blog 28 35 80.00%
4 Ocarina Networks Carter George & Sunshine Mugrabi 20 50 40.00%
5 Sepaton Jay Livens 19 87 21.84%
6 3Par Mark Farley 66 614 10.75%
7 Xiotech Xiotech blog 24 300 8.00%
8 Pillar Mike Workman 17 500 3.40%
9 FalconStor Chris Poelker 13 505 2.57%
10 HDS Hu Yoshida 67 2700 2.48%
11 NetApp Val Bercovici 107 7976 1.34%
12 HDS Michael Hay 36 2700 1.33%
13 HDS David Merril 36 2700 1.33%
14 NetApp Vaughn Stewart 103 7976 1.29%
15 NetApp Dave Hitz 95 7976 1.19%
16 HDS Pete Gerr 30 2700 1.11%
17 NetApp Nick Triantos 63 7976 0.79%
18 NetApp Alex McDonald 56 7976 0.70%
19 NetApp Larry Freeman 48 7976 0.60%
20 EMC Chuck Hollis 205 42000 0.49%
21 NetApp Storage Efficiency 36 7976 0.45%
22 EMC Mark Twomey 182 42000 0.43%
23 EMC Barry Burke 175 42000 0.42%
24 EMC Dave Graham 153 42000 0.36%
25 HP StorageWorks 99 321000 0.03%
26 Sun Brendan Gregg 6 27596 0.02%
27 Sun Adam Leventhal 4 27596 0.014%
28 IBM Barry Whyte 41 398455 0.010%

StorageMonkeys ranked the Storage Blogs based on the number of votes. We all know that a company with tens of thousands of employees should have more leverage to bring out the vote when compared to companies with considerably less resources. Cleversafe’s ranking calculation is based on the percentage of votes compared to number of employees in the company.

Because at the end of the day, isn’t it all about how you slice the data?

The real question is do these blogs have great content? Here’s a few links for recent posts – you decide.
Major Trends at StorageVisions 2010
Silent Errors
Trends in the Advancement of Storage Virtualization

* Employee numbers gathered through the internet and are not guaranteed to be accurate as of today.

Major Trends at Storage Visions 2010

By Kumar Abhijeet, Business Development Manager, Cleversafe

I was at the Storage Visions conference earlier this month and sat through multiple sessions. Overall, I came away from the conference with three emerging trends:

  1. Storage requirements in the Media and Entertainment market are growing and different from enterprise storage
  2. The unsolved problem in long term retention and archiving of data is for the storage system to  protect the data from latent failures and operational failures of drives
  3. Cloud storage is moving from just an idea for organizations, towards  real implementations

The first trend is the new storage requirements that are emerging within the Media and Entertainment market segment. If you look at this segment, the market as a whole is moving from SD to HD to 3D HD to 3D-4K. Storage has to manage more and more files, the availability of the system has to be greater than 5 nine’s, archives need to be maintained at the highest resolution( which are typically 20x larger), production times are shrinking, workflows are going tapeless and early films are being digitized and restored at high resolutions. All these trends translate to a good thing for the storage companies.

Conversely, typical enterprise storage requirements are very different from the requirements of rich media content within the Media and Entertainment segment. Enterprise hot buttons are Deduplication, Heterogeneous ILM, Cloud storage etc. Rich media content requires performance for 2K and 4K capture (368MBps per stream, often multiple streams), instant access to archive data (eliminating tape for news, TV episodes, sports, pre-recorded live), graphic representation of millions of video and audio assets and repurposing duplicated assets in various resolution and formats. When we look at these requirements against Cleversafe dispersed storage capabilities, we believe Cleversafe’s dispersed storage fits very well within the storage workflow as a nearline storage to the front end applications such as AVID, Final Cut Pro, Sienna TV, Artesia etc. Media asset management application such as Front Porch Digital and Tedial are also going to be an important piece of the work flow. In order to meet the ingest and retrieve throughput requirements there is definitely a play for Flash and SSD in front of a reliable, secure and cost effective nearline geographically dispersed storage cloud. The geo-dispersed nature of dispersed storage provides protection from data loss and provides continuity of operation even when a site goes down without the need for replication.

The second trend I noticed was the discussion around the unsolved problem in long term retention of mission-critical data. SNIA 100 year task force recently released 100 year archive requirements survey report which highlighted that long-term digital information retention are real. 68% of respondents said they keep the data for more than 100 years and  about 80% of respondents have information they must keep for more than 50 years and 70% of respondents say they are ‘highly dissatisfied’ with their expected ability to read their retained information in 50 years. The goals of digital preservation are that digital content should remain accessible, usable, and undamaged for as long as desired and beyond the lifetime of any storage technology at an affordable cost. All disk-based storage systems are going to be facing a challenge of how to protect large amounts of data against latent failures. Disk capacity is becoming larger and larger according to Moore’s law (we already have 2TB drives coming out), but Unrecoverable read errors (URE) is fixed at about 1014, meaning once every 1014 bits, there will be a bit that is unrecoverable. Doesn’t seem significant? In today’s large storage systems, it is. Unfortunately, the likelihood of having one drive fail, and encountering a bit rate error when rebuilding from the remaining RAID set is highly probable in real world scenarios. To put this into perspective, when reading 10 terabytes, the probability of an unreadable bit is likely (56%), and when reading 100 terabytes, it is nearly certain (99.97%). Cleversafe is the leader in protecting the stored data from data loss or corruption due to URE’s, disk failures and correlated disk failures by in implementing advanced erasure codes in storage software to protect from data loss. My colleague has written a very detailed blog on how Cleversafe protects the data from latent failures: http://dev.cleversafe.org/weblog/?p=259

The third trend is around the reality of cloud storage – and this is what got me really excited. Come to think of it, unlike Internet or Web2.0 based applications, cloud storage is one of the services for which end consumers, small, medium and large enterprises are actually spending money. There were a lot of numbers thrown around during the conference, but according to iSuppli, the cloud storage market will triple in three years from $1.6B in 2009 to about $5B in 2013. Amazon is the leader in providing storage services today, but when the telco operators such as AT&T, Verizon, Comcast, Savvis and others enter the cloud storage space, this will be an exciting space to be in. There are a multitude of vendors that are emerging in this space. EMC Atmos and Cleversafe are emerging as end to end cloud storage platform vendors. Parascale and Mezeo are emerging as software only vendors. I participated in a panel discussion on this topic along with IBM, Bycast and Asankya. The panel was hosted by Robin Harris. Robin has published his take on his storagemojo blog: http://storagemojo.com/2010/01/13/cloud-at-storage-visions-2010

Overall, I did not see a cloud storage provider who provides protection from data loss without replicating data objects. This is where Cleversafe fits very well. Cleversafe’s dispersed storage provides the data protection without the need for replication, hence reducing cost for the service provider and end consumers. Cleversafe’s object based dispersed storage technology provides for a platform that addresses the challenges of cloud storage: scalability, privacy, data protection, system uptime, data security / privacy and easy manageability.

Keep an eye out for the cloud storage market to evolve further in 2010. And I am excited to be a part of this evolution.

SearchStorage – top 10 enterprise data storage news stories of 2009

SearchStorage’s Beth Pariseau recently posted the Top 10 enterprise data storage news stories of 2009 . They were:

10. Data deduplication branches out
9. Object-based data storage re-invented
8. Solid-state drives hit market, but adoption slow
7. The NAS renaissance
6. Big vendors stack up
5. Clouds everywhere
4. vSphere 4 adds long-awaited data storage features
3. The IBM/Sun/Oracle love triangle
2. EMC and NetApp in bidding war for Data Domain
1. The economy

Here are some thoughts on how numero 9, and why object storage is being reinvented…

Provisioning for virtualized storage

Virtualized storage promises shared storage resources that can be parceled out as needed. One of the downsides to block based storage is that when a LUN is provisioned, the disk drives are actually formatted to accommodate the provisioned space. This leads to inefficiencies because often LUNs are underutilized, and having to earmark the space up front leads storage administrators to over provision since LUNs cannot grow is size. Thin provisioning solutions address this problem, but, are basically a layer of abstraction on top of block.

With object storage, the storage containers can be provisioned without having to actually allocate physical disk. This means thin provisioning is inherent to the storage mechanism, not an add on. As private and public cloud storage (yes a trendy word, but it does help communicate the point) emerges, being able to virtually provision is key to making these systems efficient and scalable.

Data mobility

With object storage, a unique object ID is assigned to each object. This means that the objects can actually move across physical hardware without causing file references to get out of sync. The benefit is data becomes  more mobile which will is an important requirement with automated tiering, and hardware refreshes in large scale systems. Data migration will become much simplier to accomplish with objects.

Scale without file system constraints

Imagine listing out the file system for google? Not doable. Furthermore, not valuable for end users. Google has proven that searches that show instantaneous results of relevant information are more important than traversing a file system folder structure to search for data.

Here’s another post related to Cleversafe’s object storage release from September that has further discussion.