Feeds:
Posts
Comments

Today we announced that one of our customers, Shopzilla, has been selected as a Best Practices in Storage award finalist and will be honored at the upcoming SNW event.  It’s very exciting to have one of our customers recognized for their outstanding work, but it’s also significant validation of the impact real-time compression is having in reducing storage costs and enhancing storage efficiency.  Here’s a little more detail about their deployment.

Shopzilla is a comparison shopping engine serving approximately 19 million unique visitors per month.  Shopzilla’s secret sauce is its ability to refresh more than 40 million products, six times per day while its nearest competitors can only refresh their product offerings once every 3-4 days.  Its two main data centers each have thousands of servers with several hundred TB’s per location. Shopzilla HQ supports these locations with more than 400 test, development, and Q/A machines.

Facing 200% data growth, increasing costs and ever-growing power consumption, Shopzilla decided to implement Storwize real-time compression.  Based upon Storwize’s highly efficient, lossless data compression algorithms to optimize online file-based storage, Shopzilla has experienced > 50% reduction in storage capacity requirements AND increased storage performance and efficiency.

The reduction in storage capacity requirements also delivered significant financial benefits including over $375,000 in savings, rapid ROI, and an internal rate of return of 700%.

To learn more about the details of Shopzilla’s deployment, check out this Wikibon write-up.  For anyone looking to reduce storage costs without compromising performance, it’s a must read.

Congratulations to Shopzilla!

History buffs will know I’m paraphrasing the famous quote attributed to Senator Everett Dirksen, “A billion here, a billion there, and pretty soon you’re talking about real money.”   As the market leader in online data reduction our solutions are deployed in hundreds of locations around the world helping customers to optimize many petabytes of storage and dramatically reduce their storage costs.  Here are a few real-world examples of our customers achieving real-savings:

Shopzilla

As highlighted in a recent Wikibon Peer Incite session, Shopzilla is a comparison shopping engine serving approximately 19 million unique visitors per month.  Shopzilla’s data centers house thousands of servers with several hundred TB’s of storage. Facing 200% data growth, increasing costs and ever-growing power consumption, Shopzilla deployed Storwize.  With over 50% reduction in online data footprint Shopzilla is saving almost $400,000 in CAPEX and OPEX.

Check out this 5-minute podcast or this detailed ROI analysis of the Shopzilla deployment to learn more.

Leading Records Management Company

With a goal of increasing efficiency and reducing the cost of their multi-petabyte NFS file-store, this leading records management company deployed Storwize.  With almost 60% reduction in online data footprint they’re saving more than $10,000,000 over three years!

As Everett Dirksen would say ‘that’s real money!’  How much could you be saving?  Check out www.storwize.com.

This week we announced a partnership with Hitachi Data Systems  (HDS) to offer our real-time compression solutions to HDS customers.  This partnership is further validation that real-time compression has emerged as the right technology for online (non-backup) data reduction where there is limited redundant data and file access is typically random.

Storwize and HDS offer a great combination of performance and data reduction.  Storwize has been tested and validated with the Hitachi NAS Platform including the existing Hitachi NAS 3100 and 3200 as well as the recently announced Hitachi NAS 3080 and 3090 models.  According to Marc Trimuschat, Senior Director, Worlwide Technology Alliances at HDS:

“Hitachi Data Systems customers want to better utilize their storage capacity across multiple tiers and applications to meet the continued growth of file-base data.  Storwize’s real time compression technology complements the scalability and unique tiering capabilities of the Hitachi NAS Platform. Together, we enable customers to lower their storage TCO by maximizing the utilization across storage tiers.”

That’s a terrific endorsement and we’re excited to have HDS as one of our strategic partners.  Together, Storwize and HDS are addressing the rapidly emerging market for online data reduction by offering an unbeatable combination of industry-leading storage performance, scalability, and real-time data reduction that delivers capital and operational savings across all online storage tiers without compromising application performance.

Our collaboration with HDS adds to our already significant activities with our strategic partners.  Check out what we’re up to at www.storwize.com.

According to research from ESG, more than one-third of all organizations that have implemented server virtualization have seen an increase in the total amount of data they need to store and protect.  So, while VMware is a great platform for saving money on server infrastructure, it moves the cost and complexity challenge to storage potentially negating server savings.

Storwize solves the storage problem by bringing to storage what VMware brings to servers – massive foot print reduction, lower power and cooling costs, and increased infrastructure efficiency.   Storwize solutions for VMware are unlike any other in several ways:

  • Performance - Storwize real-time compression provides the best data reduction for online application data types WITHOUT COMPROMISING performance, throughput and accessibility requirements.
  • Simplicity – Storwize is simple to deploy and transparent in operation with NO required changes to applications, servers, or storage.
  • Storage Independence – Storwize delivers data reduction savings on any NAS vendor’s storage.
  • Efficiency - Storwize is a dedicated appliance that reduces the size of files before they are written to disk which translates into more efficient filer disk and CPU utilization, shorter backup and restore times, and lower replication costs. 

Helphire, the UK’s market-leading accident management company, illustrates perfectly the benefits of deploying Storwize and VMware.  They needed to increase IT infrastructure efficiency but after deploying VMware they were faced with a challenging reality. VMware provided substantial benefits but had a significant impact on production capacity requirements, to the extent that the savings gained by server virtualization were significantly reduced by the need to purchase additional disks.

Storwize solved their problem by delivering 65% data reduction and enhanced storage peformance.  Using a combination of both VMware and Storwize enabled Helphire to control expansion costs of their network storage infrastructure while dramatically minimizing the need for power, cooling and floor space in their data center!

If you’d like to see our solutions in action, come see us at VMworld.  We’re in booth 1932.  Hope to see you there.

Steve Duplessie recently wrote in his latest Steve’s IT Rants about Scarcity Imbalances based upon the book ”Free” by Chris Anderson.   The point is simple; when things are scarce in terms of supply and demand, value increases.  In the context of storage, I’m sure he ruffled some feathers by concluding that the capacity economics of deduplication are no longer valid, but as usual he makes a strong argument that capacity is not scarce.  For example, if a dedupe solution is $90,000 for 30TB but you can purchase 30TB capacity for $10,000, why buy deduplication?  There must be compelling scarcity factors other than capacity economics, right?  The answer is absolutely yes and the reasons all revolve around operational scarcity factors.  Let’s look briefly at the impact of data reduction on some of the most significant scarcity factors:

  • Floor Space Scarcity – many businessess don’t have any more floor space.  Even if they have the budget to buy capacity where are they going to put it?  Businesses need to store and protect more with the infrastructure they already have.
  • Power & Cooling Scarcity – lack of power and cooling is certainly a scarcity factor.  Who doesn’t know someone with equipment sitting in their data center that they can’t turn on because they don’t have enough power?  There’s huge value in being able to scale a data center without requiring more power and cooling.
  • Time Scarcity- time will alway be scarce.  Time can be measured in any number of operational metrics – time to backup, time to restore, time to replicate, time to migrate, time to do a build, time to manage, etc.  The ability to change any of these metrics can have a direct economic impact on the bottom line.
  • Economic Scarcity – given today’s economic realities, businesses must figure out how to do more with less.  So, while infrastructure assets like capacity and bandwidth may not be in scarce supply businesses are desperate to leverage them more efficiently.  For example, why buy more capacity if you can get more capacity from what you already have on the floor?  Why purchase bandwidth to support replication of 10TB of data/day if you can get away with only replicating 1TB/day?

The bottom line is that data reduction is simply an enabling technology; the real economic value is in enabling organizations to better leverage their scarce operational assets to manage and grow their business.  Changing that dynamic changes the economics of IT, and for our part that’s what Storwize is doing with real-time compression.

To add some color to my point, I refer to some recent Gartner research.  Online data compression was recently highlighted in Gartner’s Hype Cycle for Storage Hardware Technologies, 2009.  In the report, Dave Russell, Gartner Research Vice President comments; “The positive impact of high data compression ratios on the need for additional storage purchases, new application deployments, operations, facility requirements and environmental costs will change the design of storage infrastructures, as well as backup/restore and archiving solutions.”

So, what are you really buying with online data reduction?  You’re buying leverage.  Leverage to do much more with the scarce assets you have at your disposal.

In a recent post, Storagebod highlights several use cases where two of today’s ‘hot’ storage management tools, thin provisioning and deduplication, may be more disruptive than beneficial.  For the purpose of this blog I’ll focus on deduplication.  I’m paraphrasing but here’s the gist of Storagebod’s concerns:

  • How do you handle de-duped primary storage where you could be many times over-subscribed with de-duped storage?
  • How do you manage migrating deduped storage between arrays?  In Storagebod’s words ‘Deduping primary storage may well be one of the ultimate vendor lock-ins if we are not careful.’
  • What happens when a user decides to turn on encryption and what was many deduped copies of the same data actually becomes many full copies of the same data and you have run out of storage space in a spectacular fashion.

Each of these is a legitimate concern and speaks to the potential pitfalls of deploying deduplication for online data.  So, why not avoid these pitfalls altogether by applying real-time compression, which unlike deduplication:

  • Provides better data reduction for online, non-backup data sets.
  • Provides direct random access to compressed data so there is no re-hydration, there’s no performance degradation and you’re never over-subscribed.
  • Is completely complementary with existing migration, replication, and storage management practices so there’s no vendor lock-in.
  • Is completely complementary with downstream data reduction, backup, and encryption deployments.

In other words, everything you did before you deployed real-time compression you can do afterwards, just faster and with less infrastructure because your data set is anywhere from 2x to 10x smaller.

Deduplication is a great technology…for backup, but it does have drawbacks for online data particularly performance and co-habitation with other storage management best practices.

Prospective buyers need to understand these practicalities or run the risk of eliminating any potential data reduction ROI due to increased operational complexity and costs.

With file capacity projected to grow at 55% annually from 2008 to 2012, according to research from the ESG, economic pressures and file data growth are driving rapid adoption of data reduction technologies.  Businesses are asking for solutions that address data reduction across all tiers of storage from primary thru backup but they need help sorting through all the options and FUD in the marketplace.  This creates tremendous opportunity for channel partners looking to provide their customers with unique value by helping them address this challenge.

From our inception, Storwize has invested in the channel and we continue to do so today with the announcement of our new partner program, Partnerwize.   Building on the success of our existing channel initiatives, the new program reflects our increased investment in infrastructure and sales enablement, including:

  • Enhanced Partner Portal – the new portal provides partners a ‘one-stop shop’ for all their sales enablement needs including sales tools, deal registration, branded marketing content, as well as product and competitive information.
  • Education Programs – the new program includes programs to educate both partners and their customers on online data reduction technology as well as the business and operational impacts of the technology.
  • Pre- & Post-Sales Support – partners have ready access to sales and technical support throughout the entire sales cycle ensuring the best customer experience.

Online data reduction is rapidly emerging as a must-have technology.  In the same ESG research only 15% of respondents said they would buy NAS without data reduction.  We’re excited to be working with our strategy channel partners to meet this need.  Are the rest of you ready to carpe diem?

Symantec recently issued a press release announcing the next phase of their deduplication strategy.  Symantec quotes a recent study that states “51% of organizations expect to spend more on storage in 2009 even though more than 90% have lower budgets and fewer people than in 2008.”  This simply means that organizations are spending a higher percentage of their precious budget on storage.  A strong argument for data reduction!  Symantec also correctly points out that, “Today’s deduplication appliances have limited impact since they only address the end of the information management lifecycle.”  This is a GREAT point and a message we’ve been preaching for awhile.  Customers need data reduction that addresses the entire data lifecycle.

Symantec then recommends ”organizations put storage on a diet by deploying deduplication technology to curb data bloat and save costs.”  But is deduplication really a diet or a purge?  In order to dedupe you’ve got to consume.  Who wants to eat and then purge anything they’ve already eaten?  Yuck.

In storage terms, why go to all that trouble of storing all the data if you’re going to ‘throw away’ (via lossless data reduction) a high percentage of the data before it’s even accessed?  Why not design your storage infrastructure to retain only the reduced data set  worth keeping?  Enter real-time compression, which reduces your data set on-the-fly BEFORE it is written to disk without even bothering to store the rest. It’s like a magical diet: You get to eat the whole meal, but you only digest the nutritious parts!

It is great to see Symantec expanding the scope of data reduction outside of backup, but it’s also important to apply the right technology to solve the problem.  For non-backup applications real-time compression is that technology.

Dave Vellante wrote a very interesting article on Wikibon.org entitled ‘The Pitfalls of Compressing Online Storage.’  In case you’re not familiar, Wikibon is ’a peer professional community solving technology and business problems through an open source sharing of free advisory knowledge.’  In other words it’s a great place to avoid marketing hype and learn from IT practitioners about the benefits and pitfalls of deploying various technologies.  Dave’s article is based upon a recent Peer Incite session where Shopzilla, UBS, and others discussed their experience with online data reduction, specifically Storwize real-time compression.

Dave’s article is a must read for anyone considering data reduction because it provides an unbiased view of the realities of deploying data reduction in primary storage.  The bottom line is that there is a lot of hype and prospective buyers must educate themselves on the various technologies for online data reduction including array-based deduplication, post-process primary deduplication, and inline compression.  After illustrating the potential drawbacks of these techologies Dave concludes that ‘Storewize will achieve the highest levels of compression and ROI in general purpose enterprise data centers.’

In the addition to the article, there are a number of valuable resources to review including a 5-minute podcast featuring Shopzilla and UBS commentary, as well as a detailed ROI analysis of the Shopzilla deployment.

Online data reduction is becoming a must have for organizations looking to reduce storage costs and enhance the performance and efficiency of their WAN, backup, and DR infrastructure, so it’s critically important prospective buyers educate themselves and apply the right technology to avoid creating more challenges than benefits.

As we’ve all heard by now Data Domain has entered into a definitive agreement to be acquired by NetApp for ~$1.5B, a 40% premium over Data Domain’s closing price yesterday.  Clearly NetApp sees significant market opportunity but it’s also clear validation that data reduction is not a simple feature that anybody can develop.  If it was why would NetApp pay close to 30% of their current valuation to purchase Data Domain.  NetApp has sent the clearest message to-date that data reduction (compression and deduplication) is now a core pillar of their storage efficiency strategy.  

NetApp’s acquisition also sends a clear message to all the other storage and network optimization vendors including HP, IBM, Dell, EMC, and Riverbed who have been touting their own data reduction solutions, current or planned.  NetApp’s message; If you want to be competitive you’ll need the IP, technology, and solutions to deliver data reduction that optimizes all tiers of storage, from primary to backup.  In an interview today Jay Kidd, NetApp CMO stated, “The technical approaches are slightly different, but their solution deduplicates backup data extremely well. NetApp dedupes primary data extremely well. The two workloads are quite different.”  At Storwize, we couldn’t agree more, the workloads are very different. Now, it will be interesting to watch how the other technology vendors respond.

This is exciting news for Storwize.  Our focus has always been on primary storage data reduction (vs. backup data reduction) and we’ve established ourselves as the market leader by proving our technology at hundreds of locations within leading companies around the world.  We support all leading NAS vendors and our unique real-time compression technology has the distinct advantage of providing significant data reduction for any primary storage application without compromising application performance; something no deduplication vendor can claim. And customers can use our solutions in conjunction with deduplication to maximize savings throughout the lifecycle. To learn more about that check out our Storwize/NetApp and Storwize/Data Domain solution whitepapers.

Congratulations to NetApp and Data Domain.  This is exciting news for the industry and excellent news for customers as you’ve raised the comptitive bar for data reduction.  It will be interesting so see how things unfold over the coming weeks and months.

Older Posts »