Tag:

HyperScale

NetApp Insight 2016 – General Session – Day 1

by Roger Lund September 26, 2016

written by Roger Lund

Today I am joining you live from NetApp Insight 2016.

This is my best attempt to keep up with the keynote as it happens, Pardon any missed content.
Roger Anderson joins us on stage after a nice media presentation, “who said data wasn’t exciting”

He talks about the value to both Customers and partners, and that they have made this event something special.

And welcomes us to #netappinsight 2016.

The conversation turns to the event, and how it is a netapp milestone, and now with feedback, less general sessions, more breakout sessions and labs than every before.

Roger Introduced George Kurian. He Talks about Digital Transformation, and that Data is the currency of the digital economy.

He goes on to talk about Netapp’s Success in Data Management in the last 20 years.

He brings up Simplify, And Modernize IT in Infrastracture , Scale and Accelerating Business Performance. Examples are SSD, Scale out, and Solidfire.

He talks about Data Silos , and how they Hinder the Data Center, and how the competition focuses on technology that hinders transformation. Netapp’s Data Fabric Enables IT’s Digital Transformation.

He goes into Transforming Netapp to serve you in the New Era of IT. by Serving more device customers with expanding technology innovation.

Next George is Addressing how customers want to buy and consume solutions by embracing the partner ecosystems.

He indicates that Netapp is the fastest Growing San vendor.

He talks about More and More Customers and Services Providers use Cloud because of Netapp Cloud Strategies.

He Introduces the 2016 Innovation Award Finalists. Mercy Technology services took the stage and talked about the success they have had with Netapp and SSD.

He leaves us with one thought. Challenge us to to make us better. And Welcome to Netapp Insight.

George introduces Joel Reich. Joel talks about yesterday’s Data Management, and how it has changed in the last 50 years.

He talks about Data, and the Power of it. Data is the new oil. The Netapp Data Fabric, is the new bank.

The solution to catching in on your new data, is data fabric.

Joe introduces the new all flash FAS AFF A700, and AFF A300. New Hybrid Models FAS9000, FAS8200, FAS 2600.

Joe Shows off the Netapp Flash Platform, with EF, AFF, And Solidfire. Next they plan a video on Solidfire.

Joe talks about Oncommand Insight and the power of Analytics.

Next they close with a demo of Snap Center, Snap Mirror, with AltaVault to the cloud. Next they backup to Storage Grid, and then Azure. Then they show how they can use Data Fabric to restore files from Amazon to your Data Center.

Next they talk about Data in Space, and NASA, and the JPL Data Center. They talk about how they use data transformation from file to object and back, with security. They use Cloud Sync, and do a demo. This allows File to Object, and back Object to file.

Cloud Sync is onthe Amazon Market Place.

Now Henri Richard takes the stage.

He talks about Partnerships, and Introduces Samsung. Samsung talks about trends and the 2.2 connections for each human on the planet. He is talking about the massive amount of data these devices are making.

Samsung talking about 3d NAN, 16TB SSD, and the newly announced 32TB SSD’s. They talk about Z-NAD Z-SSD.

Netapp AFF8020, and Multi-Stream , Custom Firmware for Improved NAND, Endurance.

He closes talking about the History of Samsung and Netapp Innovation.

Next up is Cisco. They play a video featuring Flexpod. Cisco Takes the stage, and they talk about the partnership, and flexpod. How they have 7300 flexpod customers, 1100 partners, and more.

Next they discuss the Digital Transformation, and the driving changes in IT. They detail how Application Architectures , Operational Models, and Consumption models need to change , and we are living those shifts today.

Next Cisco talks about the Data Center Strategy, and how they have making strides with a Programable infrastructure , policy driven platform. They talk about Solidfire and Flexpod, and overall customer simplicity.

Next up is Fujitsu. They talk about the Human centric innovation.

Roger Lund

September 26, 2016 0 comments

Datacenter News Storage

Atlantis Computing Expands HyperScale Lineup

by Michael Davis January 16, 2016

written by Michael Davis

Atlantis Computing has expanded their hyperconverged lineup with the HyperScale CX-4; a new offering targeting both ROBO deployments and smaller SMB customers. The two node CX-4 features 4TB (usable capacity) of all flash storage, 10GbE networking, 48 CPU cores, and up to 1.5TB of RAM in a tidy 2U package starting at $43,000 with three years of 24/7 support (4 hour parts replacement). This price point places it among the lowest cost hyperconverged appliances on the market with Chetan Venkatesh, Atlantis CEO, calling it a game changer for ROBO customers.

To make deployment easier and faster if a customer provides the IPs Atlantis will ship units with all the networking pre-configured along with connection details for the Atlantis Manager virtual appliance whether running centrally or on the ROBO appliance. Additionally the check boxes of inclusive data protection, high availability and disaster recovery are all ticked.

The expanded HyperScale lineup now includes the CX-4, CX-12, and CX-24 which are available on Dell’s FX2 modular server platform. This marks the first time Dell’s FX2 platform has been used for a hyperconverged appliance and allows for 10GbE networking between nodes without a separate 10GbE switch. Dell based CX series appliances are available through Dell distributors and channel partners in the US, Europe and Middle East. Fans of Lenovo, Cisco, SuperMicro, and HP will be pleased to see the entire lineup available from those manufacturers as well.

For more details and to configure your own, head over to Atlantis Computing.

January 16, 2016 0 comments

Compute Datacenter Storage Trends

HP shares Big Data directions

by Philip Sellers September 16, 2013

written by Philip Sellers

bigdatalarge A couple weeks ago, I had the opportunity to attend a Big Data briefing from HP Chief Technologist for Data Management, Greg Battas. Battas is part of the newly formed Converged Systems group in HP. He was a pioneer of Very Large Databases (VLDBs) and analytics in the telecom industry who has worked in business intelligence and analytics for a couple decades. He speaks internationally on topics of data integration and holds several patents in the areas of Relations Database, parallel query optimization and real time infrastructure architectures.

Coming from a mid market company, Big Data seems like a problem that doesn’t affect me. Its a concept I have a difficult time wrapping my head around, and for that reason, I’ve not written about it in the past. But it seems that Big Data is more than just a buzzword of the year and there is a lot of innovation occurring around Big Data in attempts to solve customer problems with these datasets.

So, what is Big Data? Wikipedia seems to sum it up best by saying “Big data is the term for a collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools or traditional data processing applications.” That helps.

So within the realm of Big Data, you begin to think about supercomputers, hyper-scale arrays of servers and new innovations like in-memory databases. So where is this going and what’s happening in Big Data today?

Direction: Consolidation of Clusters

Big Data Architectures in the past were based around very simple servers with direct attached storage. It leads to a first Hadoop cluster, then a second, third and fourth. A lot of the thought in Big Data was movement away from proprietary storage and databases with parallel programming and distributed file systems like Google File System on standard hardware. Common wisdom dictates that big data depends on massive IO to read huge amounts of data from disk and to optimize this you move compute closer to the data to reduce overhead. There are certain operations that benefit from being pushed down to the same location as the data.

In reality, what has been learned from Big Data workloads is that only a portion of the processing can be close to the data. Keeping data and compute local to one another is actually difficult and does not always result in optimal performance because the data may not be in the appropriate form for processing and may need to be shuffled or reduced. In addition, its observed that the majority of CPU power is still needed for analytics and aggregation. And when most work is done in-memory, the storage really doesn’t apply. This becomes particularly important with NVRAM and other persistent RAM technologies in the future. What all this allows is for the consolidation and re-tasking of clusters to meet multiple needs.

Direction: Software Defined Storage

Big Data workloads, particularly the largest datasets in the world, are running on distributed file systems on industry standard server hardware. These are parallel file systems rather than traditional storage arrays or databases. Hadoop Distributed File System is becoming a very common interface across multiple platforms and deployments and vendors are adopting HDFS and integrating it under their technologies. Today, there is a mix of proprietary and open source technologies.

HP is observing a lot of vendors running their proprietary systems on top of the HDFS, or a very similar parallel filesystem. HP’s internal blueprint with HAVEn is much the same. HAVEn has the unstructured data of Autonomy, the structured analytics of Vertica and running it on top of the HDFS. In addition to storage, the HDFS allows for data to be passed from one tool to another.

Direction: NoSQL is being adopted by a lot of software partners

The first wave of Big Data was around Batch where Hadoop was used for analytics and ETL offloading. It was often coupled with a company’s SQL databases. There is a growing interest in NoSQL products from independent software vendors because of a large dependence on the traditional database vendors. ISVs are seeing a large portion of their total sales being directed back to a database vendor at the close of a sale, so many are beginning to move their commercial products onto NoSQL products that do not require the large costs. It isn’t without challenges, since there is no longer a SQL interface or transaction management to move onto a NoSQL product that no longer requires the large costs of traditional databases. But there are several with active projects porting their products onto a NoSQL product like Hbase.

Direction: Shift to optimized hardware

With a specialized workload like Big Data, there is opportunity for tuning and tailoring hardware to specifically handle the work better than industry standard hardware. Within HP, Moonshot is one example of hardware that is hyper-scale, simple node architecture suited for Big Data (I previously covered Moonshot here). On each cartidge, co-processors or GPU’s could be added to better handle big data workloads, but there is even possibilities within the system-on-a-chip on the cartridge.

Battas mentioned the idea of Dark Silicon, or un-used and un-powered transistors within today’s chips. Today, the industry has the capability of packing on more transistors into a piece of silicon than we have connections to power, leaving them dark. But the interesting idea is tailoring the dark silicon cores to handle a specific task well and then rotating the customized core in and out of use to increase efficiency. This is a particularly interesting topic, enough so that I have written a second post today about dark silicon.

September 16, 2013 0 comments

Compute Datacenter Trends

HP and chipmakers exploring use of specialized dark silicon for improved performance and efficiency

by Philip Sellers September 16, 2013

written by Philip Sellers

I love when a briefing or conference call sends me searching for a new term I don’t recognize. I had that happen during an HP briefing with Greg Battas about Big Data last month. It was the first time I had heard of the concept of Dark Silicon, where there are more transistors on a chip than can be powered leaving some of them dark. To me, it sounded a lot like dark fiber – fiber laid in the ground but unlit because there is no use for it at the time.

When I began digging into Dark Silicon, there were relatively few sites with any information about it. Most sites that referred to Dark Silicon were from 2010 and 2011 and were referring to the pace as which chip makers can pack transistors on a chip was quickly outpacing the ability to keep them powered. Also, as the ability to pack more transistors increases, the size of the process stays about the same, leading to more and more cores being present in the same space. In many cases, the 4 and 8-core chips common in systems today physically have more cores on them, but cannot be powered. But at the same time, the cost of omitting these transistors is negligible for the manufacturer.

From the conference call I was on, Battas relayed to us that some interesting things are happening particularly in system on a chip designs. These applications are particularly power sensitive in mobile devices leaving more dark silicon in these devices. But the emerging idea is that instead of simply leaving the additional cores dark, these could be engineered for specific tasks and powered only when needed to improve the efficiency of a task.

Specialized Cores

I was able to find a great resource on the topic – a presentation titled “Is Dark Silicon Useful” from Michael B. Taylor, an associate professor at the University of California, San Diego. The presentation has a good general explanation of dark silicon and it details the topic that Battas mentioned on the call in more detail. The slide deck is available online and I suggest that you take a look at page 40 and beyond in particular.

The slide on page 43 is extremely interesting showing a 91% savings of energy through using conservation cores – which were specialty dark silicon cores tuned for a specific task, rather than traditional RISC based processing. One of the two insights that Taylor shares about conservation cores is that “specialized logic can improve energy efficiency by 10-1000x.” That’s pretty compelling.

In terms of HP and Big Data

In April, HP launched Moonshot for the world. Moonshot is a hyperscale, datacenter solution that uses server cartridges to massive scale out software. The cartridge (think blade, but more simplistic) includes all the workings of a server while offloading the power, cooling and networking to the enclosure.

After a couple briefings that have mentioned Moonshot, a few things become clear. First, Moonshot cartridges can be introduced on short, iterative cycles allowing for faster innovation and integration of the latest technologies. Second, the cartridges can be customized to perform specific tasks. One of the early briefings mentioned creating a cartridge GPU’s integrated on the board to handle facial recognition at ATMs. There is unique flexibility in that design.

Now bringing in the idea of dark silicon on Moonshot cartridges, designers and engineers are thinking that creating silicon specifically tuned to certain processes that can turn on and turn off regions as needed will improve efficiency even further with these system on a chip designs in the datacenter. There are applications like voice service processors which can greatly benefit of custom silicon.

“Because adding transistors to an existing piece of silicon is really cost effective, it lowers the barrier for putting really interesting acceleration onto the chip for databases,” according to Battas. HP really sees a lot of this innovation occurring from system on a chip vendors who are looking at applications for database accelleration. This could have large impacts in scale out architectures like Hadoop clusters and other Big Data applications.

In addition to the chip innovations, big data software is changing also as open source alternatives to the traditional big data software. This speeds adoption of new hardware innovations into the software is accelerated and this allows for new concepts to be rapidly adopted into platforms. For instance, Intel is working closely with the Apache Hadoop project to make sure that the software can take advantage of their hardware. Battas expects to see close collaboration by other big data software vendors and hardware vendors to try and exploit these types of increased performance with specialized hardware.

September 16, 2013 1 comment

Compute Datacenter Trends

HP’s Moonshot powering its corporate website, web farms

by Philip Sellers June 13, 2013

written by Philip Sellers

HP’s Proliant Moonshot servers are powering its website, serving up 300,000 visits per month on less than it takes to power 12, 60-watt light bulbs, says executives, including Meg Whitman, during their keynotes this week at HP Discover. The Proliant Moonshot platform was officially launched to the public in April and first ships with cartridges with low-power Intel Atom processors.

According to HP Fellow Dwight Barron, they saw the next wave of low energy processors and thought how do we leverage those processors for the next wave of server computing. With Moonshot the intent is really to go after highly specialized hardware optimized for specific applications. Cartridges for Moonshot go after solutions that do one thing very well.

Barron, who serves as the Chief Technologist for HP Servers Hyperscale, also said that the market is very new and there are many possibilities that can be explored with the systems. What Moonshot allows for the company is to take new silicon and turn it into an actual cartridge product within a few weeks. The Moonshot design allows the cartridges to interface with the global management inside of the Moonshot chassis. Each new cartridge design goes through HP’s normal checks and quality assurance before being released, but it really increases the flexibility of releasing an entirely new product and keeping current with industry changes for hyperscale systems.

Asked about use cases for Moonshot and uptake, John Gromala, Director of HP Hyperscale Server Product Marketing, says that orders for Moonshot within its first few weeks on the market are already exceeding HP’s expectations. The target markets of the product are really hosting and static web server farms. Within the hosting market, Moonshot with x86 processor cartridges would provide a path to dedicated servers at the price point similar to a virtual server instance.

At the present time, the cartridges inside of a chassis are all uniform and mean to be for scale out solutions. One example that was given was around the banking industry – with many banks moving to mobile check deposits, banks may also want to include facial recognition along with the check to be deposited, but the facial recognition could be greatly improved with special silicon made onto a front-end web server. The special silicon in an industry stand server could be accomplished with a PCI expansion card, but in Moonshot, you can buckle the GPU with a low-power CPU’s in order to achieve long-term cost savings across mass scale.

While not for mass enterprise adoption, Moonshot hits the mark for large hosting providers, where energy savings and increased density have an appeal for mass deployments. Moonshot, as it evolves, is a new innovation to watch.

Disclosure: HP sponsored my trip to HP Discover, covering all travel expenses, however HP does not control the content of the posts that I write based on my attendance to the show. The thoughts, impressions and information contained in this post are my own.

June 13, 2013 0 comments

Compute Datacenter Storage Trends

HP releases Moonshot platform for high density, eco-friendly servers

by Philip Sellers April 14, 2013

written by Philip Sellers

Last Monday, HP released the new HP Moonshot platform, a new set of server products specifically built to take less space, less power and less operational cost than traditional servers in the datacenter. In Moonshot, HP is attempting to solve the issues data centers face today with finding ample power, finding space and cooling challenges.

HP Moonshot is moniker used for both the chassis and the server cards inside of the chassis. It is a similar concept to the blade servers HP and other vendors have produced, replacing the blade with a server card similar to what you’d see in a switch chassis.

HP Moonshot System The chassis itself is an oddly sized, 4.3U server enclosure and serves as the basis for the ecosystem with 45 slots for server cards and two network switches with shared power and cooling. The card based servers are a rethink of what defines a server – a small footprint system-on-a-card based around low power processors. As for space consolidation, HP says that what previously would have occupied 10 racks of 1U servers can be accommodated in a single rack with 10 Moonshot chassis.

The first set of Proliant Moonshot Servers are based around Intel Atom processors. Being x86-based, much of the world’s current server software should work with these new servers, something that will be key to adoption of the new form factor in the datacenter. In addition to these Intel Atom processors, HP also announced plans for tailored servers for specific business needs. These servers may be stocked with specific hardware for a task and future server will include ARM processors, too. The ARM based servers will likely require a lot of rewrites of applications for compatibility, but given the amount of experienced ARM developers for mobile, its not a stretch to consider server code being written on ARM.

Oddly (to me at least), HP has a requirement that all 45 server cards be identical in an enclosure according to the QuickSpecs found online for factor integrated models (thanks to Chris Wahl for that information). At launch, there is only a single Proliant Moonshot server card available. The requirement to buy a fully populated system is a clear sign that HP is targeting the Moonshot systems towards service providers and serious datacenters running lots of web applications on traditional server hardware. It is concerning for enterprise who tend to mix and match server models in blade enclosures to fit specific needs. Moonshot’s uniform requirement may limit its audiences.

Moonshot does not seem to be a play towards hardware for virtualization, but a play towards to scale-out Google and Facebook cloud architectures. Facebook has even gone as far as creating the Open Compute Project to help further its efforts to create low cost compute and storage nodes. It seems the goals of Moonshot may be in the same vein as Open Compute. Facebook’s idea is to scale out and replicate data across many nodes, using cheap and disposable server hardware to accommodate their workloads. Clearly, Moonshot’s server cards fit that bill.

Add to this the reduced consumption of power and that makes the relatively weaker processors more palatable for many buyers. Power may be the most expensive commodity for datacenters today and the reliance on electricity is only increasing year over year. Few companies have the resources or determination of Apple to power their datacenters with nearly 100% renewable power sources – like the Maiden,NC, datacenter. That means all other datacenters are at the mercy of the power grid and pricing from their local electricity providers.

HP Moonshot is definitely a new platform to watch. It will be interesting to see if it catches on as well as the blade servers and convergence that HP helped to usher into the industry.

April 14, 2013 0 comments

Compute Datacenter Storage Trends

Looking into the future of hyperscale and ARM in the datacenter

by Philip Sellers April 6, 2012

written by Philip Sellers

It is undoubted that the next wave of computing appears to be based around low energy, high performance chips like ARM processors. All of the current generation smart phones use these system on a chip designs which increasing powerful devices. They are as capable and powerful as many of our desktop computers.

The datacenter today, however, is clearly based around the x86 and x64 architecture which has supplanted all other competitive chips from the market. There are still some small markets for RISC chips, but clearly the largest part of the market is based around x86.

With all the advantages of ARM for mobile devices, it makes sense that large companies are experimenting with using these chipsets for datacenter applications. This will certainly mean new software projects to make use of the ARM architecture, but long term, the low power and high density could invent massive new computing platforms in the future.

In October, I spent some time in Houston at the HP facility there learning about their newest innovations, such as hyperscale computing and EcoPOD datacenters – all innovations seeking to increase the density and efficiency of computing. HP ‘RedStone’ development platform is a proof-of-concept hardware platform and is created to get HP’s partner ecosystem some hands-on time to run their applications and figure out what this new configuration can be potentially leveraged for. The overall strategy moving forward for these ultra low power, high density computing is what HP is calling Project Moonshot, and ‘RedStone’ is just the first rocket on that trajectory.

‘RedStone’ boasts some impressive numbers in terms of density and power consumption. A single rack unit can enclose over 3,000 system on a chip processors. The equivalent of this in an x86 architecture would take an entire row of racks for that number of processors.

Lets be honest. There aren’t many customers who can really harness a single rack with over 3,000 CPU’s. Applications for such hardware are specialized and the customers are large or specialty organizations, initially I’m thinking super computing. But, also, web farms and large scale cloud computing could be potential applications too. For large scale websites like Amazon and Facebook, adapting their software to use ARM could yield large benefits in space and power consumption. But also, cloud vendors, where we are already changing the underlying software to some degree may also make a practical place for this in the market.

Hyperscale and ‘Redstone’ are clearly plays to the largest players in the datacenter and enterprise – content farms and massive websites plus supercomputing applications. But over time, I think that even 1U and 2U rack-mount enclosures for these could trickle into the traditional datacenter, further driving convergence and shrinking footprints of compute in similar ways that blade enclosures have.

I realize that much of this is old news, but it has taken a bit of time for me to process through what this actually means to my datacenter. I initially thought these announcements had no bearing on my daily work, but in recent weeks, I have realized that any innovations made for the large players in computing would eventually trickle down to the smallest of datacenters as well.

I really began to think about how these ultra-small computers could really revolutionize small office servers and computing. In my consulting years, I worked for a number of small law offices. Most of these ran Microsoft Small Business Server for email and shared calendaring, as well as file storage and faxing. These offices would have been able to help from virtualization and failover to eliminate their single point of failure, but the required infrastructure really placed these capabilities out of their reach. But, packaging several ARM systems into a single enclosure with bundled flash storage available to all of them could easily take and extend virtualization and cloud towards this market. Even more so, since the software will need to be rewritten, it can take into account a completely different architecture during deployment.

Both VMware and Microsoft are working on ports of their flagship OS towards ARM processors – VMware experimenting with dual-personality phones and a hypervisor to separate work and pleasure on a personal device and Microsoft is writing the next version of Windows for ARM in hopes of gaining the tablet market. Once the core code is ported, its a matter of time before these can extend to datacenter specific versions. Not to mention, Linux vendors are already quickly working towards porting their OSes to ARM for use in these applications.

I’ll admit, its pretty cool stuff to think about the what-ifs and possibilities that this could lead to. I don’t have a crystal ball or any sort of insider knowledge, all the above is simply my wild imagination and speculation.

April 6, 2012 0 comments