Tag Archive for storage

Violin Memory have cool technology, but do they have a future? I hope so!

Since visiting Violin Memory at Storage Field Day 8, it has taken me a while to get around to writing this post, and I guess the reason is because I am both frustrated, and a little bit sad.

They were the first guys who (for me at least) truly tamed the beast that is flash storage, packed it up into a blisteringly fast product with insanely low latencies, and released it into the big wide world.

An organisation I worked at took that product and threw thousands of desktop users at it, a number of very busy SQL TempDBs, and some other frankly evil workloads. It didn’t even blink an eye! With Violin Storage on the back, I have seen RDS desktop customers stress test the platform with up to 100 users in a single VM!

The thing that makes me sad is that to date, Violin have not yet turned a profit.

This is not some upstart company out of the boondocks! This is a mature company, founded in 2005, who provide their product to some of the biggest enterprises in the world. Yet so far they have not managed to make a penny!

Anyway, there are other guys in the industry who are far better at financial analysis than me (for example Justin Warren did a post on this very subject just last week). So I will leave it to them try to work out why this is the case, because I like to talk tech!

Violin Memory

Looking at the latest incarnation, once again Violin have “evolutionized” [sic] something which is technically very impressive.

  • The original Violin OS has been given the boot and has been replaced by something they call Concerto OS, which blends elements of the original software with some updated features. In development is also support for a multi-controller configuration beyond the current dual, though this is obviously not GA yet.
  • The new dual controller Flash Storage Platform supports FC, iSCSI and Infiniband, as well as RDMA and ROCE.
  • It is all packaged in a 3U appliance with up to 64 redundant flash modules.
  • Thanks to their own custom backplane design, it is capable of 10-12GB/sec of throughput!
  • At 100% sustained writes they have measured 400,000 IOPSat RAID5, which is more than many of their competitors can achieve with 100% reads!It should be noted that this was on the performance model, which does not support dedupe.

All in all the new solution just screams FAST! In fact, I’m surprised they didn’t paint a red stripe down the side of the chassis!

low latency

So why on earth are Violin not ruling the AFA world right now? As a frickin cool technology with hyper speed storage, they deserve to be up there at the very least!

If I had to hazard a guess, it’s because for most companies, good is good enough.

With the smorgasbord of All Flash Arrays available today, if you don’t need latency measured in microseconds and massive IOPS/bandwidth, then you have a huge array of choices (pardon the pun). At that point features, price and support become more important than straight line drag racer performance.

If Violin want to compete with the general market whilst servicing their high-speed clients, then they need to concentrate on continuing to developing a wider range of data services and providing entry-level options to consume their products. The last thing you want to do is lose business just because you were missing a check box in an RFP…

If Violin can stop burning cash and break even, then perhaps they have a future. I for one, hope so!

If you want to catch Violin’s presentation from SFD8, check them out here:
http://techfieldday.com/appearance/violin-memory-presents-at-storage-field-day-8/

They will also be presenting again at SFD9 this week, and I’m looking forward to finding out what they plan to do next!

Further Reading
Some of the other SFD8 delegates have their own takes on the presentation we saw. Check them out here:

Disclaimer/Disclosure: My flights, accommodation, meals, etc, at Storage Field Day 8 were provided by Tech Field Day, but there was no expectation or request for me to write about any of the vendors products or services and I was not compensated in any way for my time at the event.

Pure Storage goes All Cloudy

Conversations (and arguments) are prevelant in the industry today about moving to the cloud, be it public cloud, private cloud or hybrid cloud.

People are developing into the usual fervent “religious” groups depending on their personal biases and use cases, and today Pure Storage joined the discussion by announcing a new feature for their FlashStack product. Working in partnership with Equinix, they will enable you to effectively colo a replicated FlashStack device in an Equinix DC, then use it to mount volumes via iSCSI, straight into your instances on AWS and Azure.

FlashStackHybrid.png

This is an interesting solution and having discussed this with several other guys at Pure Accelerate, the general impression is positive for two main reasons/uses:

  • Legacy solutions which depend on traditional enterprise infrastructure / storage patterns and performance.
  • As a sales enabler for Pure – Ticking checkboxes for FlashStack when”Is it cloud integrated?” comes up in an RFP.

It’s a cool idea, but personally, I am still struggling to find a really compelling use case for this, and I do slightly worry for Pure that they are providing an easy on-ramp to the public cloud for their customers. Once the compute is up there, comfort with the public cloud grows, and applications are refactored to be more cloud-compatible, the primary storage will undoubtedly follow. Pure do run the risk eroding a percentage of their customer base in the long term.

I guess only time will tell, but I hope this little gamble does not turn around to bite Pure in the rear later!

Hybrid FlashStack integration is available on Azure today, with AWS coming soon…

Disclaimer/Disclosure: My accommodation, meals and event entry to Pure Accelerate were provided by Pure Storage, and my flights were provided by Tech Field Day, but there was no expectation or request for me to write about any of the products or services and I was not compensated in any way for my time at the event.

Pure Storage – Now available in Petite

Pure Storage are probably one of the best known “All Flash” vendors in the industry today, but one of the things which has set the bar a little high for smaller organisations to get a slice of this speedy action, is the price.

Well, the good news is that for customers with smaller projects or simply smaller budgets, a Pure AFA is now within reach!

At Pure Accelerate today, along with their new FlashStack and FlashBlades, Pure announced a mini version of their ever popular M-series arrays, the FlashArray //M10.

This new array is fully featured, comes in the same form factor and with all of the same software, support, etc as the bigger models (//M20, //50, //70) but at a lower entry point of <$50k list including the first year of support. Not only that, but it is in-place non-disruptively upgradeable to the larger controller models later, all the way up to the //M70, so it is possible to buy in at this cheapest level and upgrade as business needs dictate later.

The only main differences between te //M10 and other Pure models is the lack of expansion ports on the controllers (you need at least an //M20 if you want to add shelves), and reduced compute / DRAM capacity.

m10.png

Specs are pretty much in line with the rest of the arrays in the range, with the //M10 coming in at 5TB / 10TB raw. Depending on your workloads, after dedupe and compression, this could be up to the stated useable (12.5TB/25TB). Mileage, as always, may vary! This is the perfect quantity for many use cases, including small to medium sized VDI environments, critical databases, etc. I suspect the //M10 may even find its way into some larger enterprises who’s internal processes often dictate that every project has its own budgets and its own pool of dedicated resources!

Lastly, and possibly most importantly to small businesses who may not have full time staff dedicated to managing storage, Pure’s monitoring and upgrade services are all included as well, via Pure1.

I think this is a positive step for the company and will help to engage with their customer base earlier in the organisational lifecycle, and when combined with their unique and very sticky Evergreen Storage offering, it will enable them to keep customers for life!

Disclaimer/Disclosure: My accommodation, meals and event entry to Pure Accelerate were provided by Pure Storage, and my flights were provided by Tech Field Day, but there was no expectation or request for me to write about any of the products or services and I was not compensated in any way for my time at the event.

Pure Storage Diversity – Time for the All Flash Vendor to go All File

It was only a couple of weeks ago I was saying to some colleagues that now Pure have finished with the whole IPO business, I thought they needed to diversify their portfolio a bit beyond the straight forward AFA.

I am very pleased to say they must have pre-read my mind and that’s exactly what they’ve announced today! 🙂

Not only is their new Pure FlashBlade platform designed to provide pretty much every file type (and object) you might require for your applications and users, it is also Scale Out, which is a key feature I am looking for more and more these days when researching new products for my customers.

FlashBlade.png

Not only is this a really interesting change in direction for Pure, but I see it as a pretty nifty bit of kit in and of itself! You would hope so, as Pure have been working on it in secret for the past two and a half years… 😮

For starters Pure have mixed and matched both Intel and ARM chips on every single blade, with different computational tasks being assigned to different chips, and a bit of FPGA technology thrown in for good measure. The latter being primarily used as a programmable data mover between the different elements of the blade, so as future flash technology becomes available, the FPGA can simply be re-coded instead of requiring total redesign / replacement with every generation. This will enable Pure to change out their flash as often as every 6 months in their production plants, taking maximum advantage of the falling prices in the NAND market.

This chip design was created to use the ARM processors as embedded and linked to the FPGAs, which effectively gives you a software overlay / management function, along with other low intensity, multi-threaded processes. The significant computational power of the Intel chips, particularly for single threaded workloads, rounds out the compute. From a nerdy technologists standpoint, all I can say is schweeeet!

The numbers they are suggesting are pretty impressive too! Each 4u appliance is capable of scaling out linearly with the following stats:

  • Up to 15x 8TB or 52TB blades, for a maximum of 1.6PB per 4u chassis
  • Up to 15GB/sec throughput per chassis, though I believe this is 4K 100% read, and real numbers might be around 1/3 of this.
  • 40Gbps ethernet out, with 2x 10Gbps per blade, connected to a broadcom based, custom, resilient backplane / switch layer within each chassis. Scaling to multiple chassis would require you to provide ToR switch ports for east-west traffic between chassis.
  • Overlaying this is Pure’s custom SDN code, which securely separates internal and external traffic, and uses multicast for auto-discovery of new devices.
  • Integrated DRAM and NV-RAM on every blade, along with PCIe access to the NAND.

The blades themselves look something like this:

blade.png

In terms of protocols, it will support NFSv3 out of the box on GA, with SMB and object storage over S3 supported shortly afterward. My understanding is that initial S3 support will be limited to basic commands, PUT, GET, etc, and more advanced feature support is in the pipeline. The initial release seems to be primarily targetted at the filer market, with object being the underlying architecture, but not the main event. As this support is built out later, the object offering could become more compelling.

The data itself is distributed and protected through the use of N+2 erasure coding, using however many blades are in the chassis. For example an 8 blade system would be configured as EC 6+2. As the number of blades in the system increases, garbage collection cycles are used to redistribute data to the new capacity, though I am still not 100% sure how this will work when your existing blades are almost full. The compute within each blade, however, acts independently of the storage and can access data resources across the chassis, so from the moment the additional blade is added, you have immediate access to the compute capacity for data processing.

My only query on this would be why Pure did not offer the ability to choose between Erasure Coding, which is ideal for lower performance requirements, and replicas, which would be handier for very low latency use cases? If you are putting it all on flash in the first place, instead of a hybrid model, there may be times when you want to keep that latency as low as possible.

The software platform they have design to manage this is called Elasticity, and to reduce the need to learn yet another interface, it looks very similar to the existing Pure management interfaces:

elasticity.png

A metadata engine with search functionality will be coming later, which will allow you to gain insights into the types of data you hold, and may potentially able to delve into the content of that data to find things such as social security numbers, etc. There are few details available on this at the time of writing.

As with the other Pure platforms, telemetry data is sent back to base on a regular basis, and Pure take care of all of the proactive maintenance and alerting for you. All of this data is presented through their Pure1 portal, which is pretty fully featured and intuitive.

I have to say I am genuinely surprised to see Pure come out with a solution with such completely bespoke hardware, when the entire industry is going in the direction of commodity + software, but the end result looks really promising. The sooner they can get SMB (not CIFS!) into the product the better, as this will allow them to begin competing properly with the likes of NetApp on the filer front.

As with many new products we tend to see on the market, the data services are not all there yet, but at the rate Pure do code releases, I don’t imagine it will be long before many of those RFP check boxes will be getting checked!

GA is expected during the second half of 2016.

Disclaimer/Disclosure: My accommodation, meals and event entry to Pure Accelerate were provided by Pure Storage, and my flights were provided by Tech Field Day, but there was no expectation or request for me to write about any of the products or services and I was not compensated in any way for my time at the event.

%d bloggers like this: