I don’t believe in coincidences. Yesterday we released a fact-filled press release on the vital importance and benefits of Storage Efficiency in aggregate (as the Foundation for Enterprise, Cloud & Big Data Infrastructure) on the same date as Greenpeace decided to focus high-profile attention on the environmental impact of large-scale IT (Cloud) infrastructures. True to our theme, this is a Big Blog post, but read on below to uncover the story behind these Big Savings!
In my role as part of NetApp’s office of the CTO, I have the great privilege of regularly talking to the architects and operators the largest IT infrastructures in the world. After years of doing this, I’ve observed a fascinating dichotomy.
These app architects are boldly forging new ground in technology areas such as parallel processing, distributed infrastructure, in-memory computing, continuous event processing and eventual consistency, amongst others. For businesses that depend entirely on their data, I’ve noticed revenue recognition and expense accounting rules permit them to adopt a very top line CapEx-focused infrastructure investment profile, deferring bottom-line OpEx realities until the growth of the business stabilizes.
The Big Story is Big Efficiency
OTOH the unsung operators who maintain said massive infrastructures universally lament that the TCO of these exciting new app architectures “paves a road to bankruptcy” to directly quote a few of them from companies you all know and use every day.
While I’m no tree-hugger, I have a huge amount of respect for the work Greenpeace does and the worthy causes they bravely pursue. And although the fine print of their latest report remains in question, there’s no doubt in my mind the Big Infrastructure unsustainability Greenpeace has highlighted proves they’re on to something …. you guessed it – BIG
I’ve long contended on this blog and during many of my industry keynotes, breakout sessions and panel discussions that the disturbing trends powering the build-out of MORE data centers featuring novel PUE is akin to treating the symptom of this Big Problem of our time. Conversely – designing, retrofitting where possible and managing LESS data centers via more efficient *infrastructures* treats the root cause and is the foundation of …
The Big Solution
A little over 5 years ago I began to blog via a highly popular guest post on Robin Harris’ StorageMojo blog about some (not coincidentally) big installed base data about the analysis of drive failures. Since then NetApp has embarked on one of the most ambitious Big Analytics projects outside of the Web / Social Media universe. In fact, we just implemented phase one of our Hadoop-based next-gen solution to efficiently process One Trillion (not a typo) records – and counting! You could say, at NetApp we do more than talk the current trendy Big Data talk. We also walk …
The Big Data Walk
This public disclosure of our ASUP.Next (slideshare link) project summarizes how we process real-world telemetry from our giant installed base to deliver cool services to customers while also sharing insights into the Macrocosm of Enterprise, Cloud and now Big Data IT overall. Here are some important data points to consider.
This simple bar graph above is an excellent proxy for NetApp’s growth over the past few years. Although I’m proud to highlight that our capacity shipped grew faster than revenue – proof positive that we deliver more value to customers with every passing quarter; there are other insights to discuss:
- The dominance of ATA drives. FlashCache technology has enabled us to drive customer storage costs lower by not only Virtually Tiering data simply and automatically, but also by literally shifting customer purchases to the densest and implicitly most green and cost-effective drive types of any given period.
- The rapid progression of Exabytes shipped. It took us a whole year to ship that gargantuan capacity in FY09. Only half that amount of time by FY11 and starting last quarter (Q3FY12) only 90 days to do it!
- Our own capacity disclosures confirm geometric growth of storage capacity, trending towards exponential.
Big Storage Efficiency
Drilling down a bit further on the 5 Exabyte capacity of big installed base data we’ve accumulated, we see the overwhelming adoption of mature Storage Efficiency technologies such as Snapshots and RAID-DP, but also strong adoption of relatively newer Storage Efficiency technologies such as Dedupe and Thin Provisioning. Thin replication via SnapMirror (the #1 storage replication technology by market share) continues its march towards universal adoption. If anything, adoption of our Thin (Virtual) Cloning technology shows below my expectation here, but I suspect that # will grow commensurate with our growth in hypervisor-dominated compute Clouds.
How can you take advantage of this Big Information?
If you currently own a deployed (non depreciated) storage solution, check out our V-series payback guarantee program (where we translate the 5 EB into other terms):
If you want to learn how much you can save as a new NetApp customer, then the tried and true Storage Efficiency Calculator (SECalc) is available online:
As an existing NetApp customer, you should definitely check out your custom MyAutoSupport dashboard:
Once again, I believe everything happens for a reason. There’s a revolutionary new release of Data ONTAP 8 coming that underpins the future of Big IT Efficiency and Big IT Savings. While the Big Marketing launch for that release will happen later this summer, as of *today* NetApp customers and partners can now download the GA version of this monumental Scale-out Data ONTAP milestone here. Enjoy!
And as always please feel free to share your Big feedback via a comment below.