Visualizing (and Optimizing) Cluster Performance

(Excerpt from original post on the Taneja Group News Blog)

Clusters are the scale-out way to go in today’s data center. Why not try to architect an infrastructure that can grow linearly in capacity and/or performance? Well, one problem is that operations can get quite complex especially when you start mixing workloads and tenants on the same cluster. In vanilla big data solutions everyone can compete, and not always fairly, for the same resources. This is a growing problem in production environments where big data apps are starting to underpin key business-impacting processes.

…(read the full post)

Is It Still Artificial Intelligence? Knowm Rolls Out Adaptive Machine Learning Stack

(Excerpt from original post on the Taneja Group News Blog)

When we want to start computing at biological scales and speeds – evolving today’s practical machine learning forward towards long-deferred visions of a more “artificial intelligence” – we’ll need to take advantage of new forms of hardware that transcend the strictly digital.

Digital computing infrastructure, based on switching digital bits and separating the functions of persisting data from processing, is now facing some big hurdles with Moore’s law. Even if there are a couple of magnitudes of improvement yet to be squeezed out of the traditional digital design paradigm, it has inherent limits in power consumption, scale, and speed. For example, there simply isn’t enough power available to meet the desires of those wishing to reach biological scale and density with traditional computing infrastructure, whether evolving artificial intelligence or more practically scaling machine learning to ever larger big data sets.

Knowm Inc. is pioneering a brilliant new form of computing that leverages the adaptive “learning” properties of memristive technology to not only persist data in fast memory (as others in the industry like HP are researching), but to inherently – and in one operation – calculate serious compute functions that would otherwise require the stored data to be offloaded into CPU’s, processed, and written back (taking more time and power).

The Knowm synapse, their circuit-level integrated unit of calculation and data persistence, was inspired by biological and natural world precedent. At a philosophical level this takes some deep thinking to fully appreciate the implications and opportunities, but this is no longer just a theory. Today, Knowm is announcing their “neuromemristive” solution to market supported with a full stack of  technologies – discrete chips, scalable simulators, defined low-level API’s and higher-level machine learning libraries, and a service that can help layer large quantities of Knowm synapses directly onto existing CMOS (Back End of Line or BEOL) designs.

Knowm is aiming squarely at the machine learning market, but here at Taneja Group we think the opportunity is much larger. This approach that intelligently harnesses analog hardware functions for extremely fast, cheap, dense and memory-inherent computing could represent a truly significant change and turning point for the whole computing industry.

I look forward to finding out who will take advantage of this solution first, and potentially cause a massively disruptive shift in not just machine learning, but in how all computing is done.
 

…(read the full post)

What was BIG at Hadoop Summit 2015

(Excerpt from original post on the Taneja Group News Blog)

At this month’s Hadoop Summit 2015 I noted two big trends. One was the continuing focus on Spark as an expansion of the big data analytical ecosystem, with main sponsor Hortonworks (great show by the way!) and most vendors talking about how they support, interact, or deliver Spark in addition to Hadoop’s MapReduce. The other was a very noticeable direction shifting focus from trotting out ever more gee-whiz big data use cases towards talking about how to make it all work in enterprise production environments. If you ask me, this second trend is the bigger deal for IT folks to pay attention to.

…(read the full post)

SOC, NOC, and Roll – AccelOps Converges Security and Network Ops

(Excerpt from original post on the Taneja Group News Blog)

We are seeing convergence everywhere in IT these days. AccelOps shows how convergence in systems management offers many of the same kinds of values as it does in other areas of IT – leveraged capabilities across formerly silo’d practices, simplified tasks and automation embedding best practices, and ready to roll deployment out of the box. AccelOps has tied security, compliance and network operations together into a one stop SOC and NOC “in a box”.

…(read the full post)

All Your Clouds Belong to Us! EMC Federates Virtustream

(Excerpt from original post on the Taneja Group News Blog)

Today EMC (the squared version) announced the acquisition of Virtustream, and the positioning of it as a new full EMC federation member alongside EMC II (storage, etc.), VMware, and Pivotal.  Virtustream is all about managing mission-critical production workload-hosting clouds, and has both a software business selling management layer solutions and an IaaS business as a service provider.

…(read the full post)

CTERA 5.0 Cloud Gateway Embeds File Sync and Share Services

(Excerpt from original post on the Taneja Group News Blog)

CTERA this week announced their 5.0 cloud gateway, which presents a compelling convergence play of distributed cloud-backed NAS, integral backup services, and now natively built-in enterprise-wide file syncing and sharing services and interfaces. Yep, end users get a fully secured endless NAS, local backup/restore to their files, and the global file and document collaboration they otherwise go around IT by using insecure public services. This really helps organizations address the broader trend toward the end-user driven”consumerization of IT”.

…(read the full post)

EMC Releases ViPR into the Wild – Storage Controllers for Everyone!

(Excerpt from original post on the Taneja Group News Blog)

At EMC World 2015 this year EMC made a bunch of announcements, but one of the most discussed, and causing the most head-scratching, was the open sourcing of their ViPR storage controller as the new Project CoprHD. Why would EMC give away the core automation component to their much bandied about “Platform 3” strategy?

…(read the full post)

Massive IOPS with Drops In Price-Performance – Early 2015 SPC-2 Standings

(Excerpt from original post on the Taneja Group News Blog)

There is something quite interesting happening on the SPC-2 top 10 results page. Right now, HP (with XP7) and Kaminario, a hot all-flash solution, are leading the performance list. But the next three entries belong to Oracle’s ZFS Storage ZS3 and ZS4 lineup. And when you scroll down a bit, the biggest surprise is that the top 10 price/performance storage leader is now Oracle!

…(read the full post)

InfiniBand Updates Specs Preparing for 10000 Node Exascale Clusters

(Excerpt from original post on the Taneja Group News Blog)

We’ve long been fans of InfiniBand, watching as new generations of enterprise class scale-out clusters and storage solutions learn from the HPC world how to achieve really high-speed interconnection. InfiniBand itself may never win the popular market race against Ethernet, but newer generations of Ethernet are looking more and more like InfiniBand. And parts of the IB world, namely RDMA and RoCE, have swept into datacenters almost unaware (e.g. look under the hood of SMB 3.0).

…(read the full post)