Fujitsu Bets On Deep Leaning And HPC Divergence

One of the luckiest coincidences in the past decade has been that the hybrid machines designed for traditional HPC simulation and modeling workloads. which combined the serial processing performance of CPUs and the parallel processing and massive memory bandwidth of GPUs, we also well suited to run machine learning training applications.

If the HPC community had not made the investments in hybrid architectures, the hyperscalers and their massive machine learning operations, which drive just about all aspects of their businesses these days, would not have seen such stellar results. (And had that not happen, many of us would have had …

Fujitsu Bets On Deep Leaning And HPC Divergence was written by Timothy Prickett Morgan at The Next Platform.

Intel, Seagate continue push for increased enterprise SSD capacity

The Flash Memory Summit is taking place in the Silicon Valley, and SSD vendors are showing off some impressive new enterprise-scale drives with greater capacity and performance.Seagate showed off some new products in its Nytro line with 3D NAND, raising capacity four-fold, and also showed off a PCI Express-based card with 64TB capacity. Conveniently, the company did not give the price. Seagate Nytro 5000 SSD line The Nytro 5000 product line is an upgrade to Seagate’s existing XM1440 line of SSDs. These drives range in capacity from 400GB to 2TB. All use the M.2 interface. M.2 is a design that’s different from traditional SATA drives. A typical SATA SSD looks like a 2.5-inch hard disk and uses a SATA interface. M.2 is about the size of a stick of gum and plugs into the motherboard. To read this article in full or to leave a comment, please click here

Intel, Seagate continue push for increased enterprise SSD capacity

The Flash Memory Summit is taking place in the Silicon Valley, and SSD vendors are showing off some impressive new enterprise-scale drives with greater capacity and performance.Seagate showed off some new products in its Nytro line with 3D NAND, raising capacity four-fold, and also showed off a PCI Express-based card with 64TB capacity. Conveniently, the company did not give the price. Seagate Nytro 5000 SSD line The Nytro 5000 product line is an upgrade to Seagate’s existing XM1440 line of SSDs. These drives range in capacity from 400GB to 2TB. All use the M.2 interface. M.2 is a design that’s different from traditional SATA drives. A typical SATA SSD looks like a 2.5-inch hard disk and uses a SATA interface. M.2 is about the size of a stick of gum and plugs into the motherboard. To read this article in full or to leave a comment, please click here

Intel, Seagate continue push for increased enterprise SSD capacity

The Flash Memory Summit is taking place in the Silicon Valley, and SSD vendors are showing off some impressive new enterprise-scale drives with greater capacity and performance.Seagate showed off some new products in its Nytro line with 3D NAND, raising capacity four-fold, and also showed off a PCI Express-based card with 64TB capacity. Conveniently, the company did not give the price. Seagate Nytro 5000 SSD line The Nytro 5000 product line is an upgrade to Seagate’s existing XM1440 line of SSDs. These drives range in capacity from 400GB to 2TB. All use the M.2 interface. M.2 is a design that’s different from traditional SATA drives. A typical SATA SSD looks like a 2.5-inch hard disk and uses a SATA interface. M.2 is about the size of a stick of gum and plugs into the motherboard. To read this article in full or to leave a comment, please click here

Worth Reading: What we learn in transitioning from card catalogs

When we transition to a new technology, do we lose valuable metadata? Could errors be introduced in the process of transfer of data from one medium to another? It’s an issue society has already had to confront — when the world’s libraries transitioned to digital and online catalogs from their older systems based on paper cards. —David Cassel @ The New Stack

The post Worth Reading: What we learn in transitioning from card catalogs appeared first on rule 11 reader.

IBM Highlights PowerAI, OpenPower System Scalability

The golden grail of deep learning has two handles. On the one hand, developing and scaling systems that can train ever-growing model sizes is one concern. And on the other side, cutting down inference latencies while preserving accuracy of trained models is another issue.

Being able to do both on the same system represents its own host of challenges, but for one group at IBM Research, focusing on the compute-intensive training element will have a performance and efficiency trickle-down effect that speed the entire deep learning workflow—from training to inference. This work, which is being led at the T.J. Watson …

IBM Highlights PowerAI, OpenPower System Scalability was written by Nicole Hemsoth at The Next Platform.

IDG Contributor Network: Driving operational excellence with your cloud vendors

Once, there was a pin factory. It employed ten workers – each of whom performed a different task. This organizational structure allowed them to generate 48,000 pins every day. If the people working at the plant were working independently, the output of each would have been limited to 20 pins at most – totaling 200 pins. This story describing division of labor was used in Adam Smith’s 1776 book The Wealth of Nations, as an example of operational excellence (OE).If your company is to survive in a competitive market, OE must be sought, explained Faisal Hoque in Fast Company. In other words, the enterprise must “identify, understand and create the capabilities, behaviors and focuses necessary for repeatable, continuous and measurable operational improvement,” said Hoque.To read this article in full or to leave a comment, please click here

On the ‘web: Hyperconvergence in the Network

First there is white box; the networking world is already moving in this direction. While network devices, such as firewalls, routers, and switches were once purchased in an “appliance” model, many parts of the networking world are quickly moving towards a disaggregated model, where the hardware and software are purchased as separate “things.” This enables the concept of white box—although the box might not actually be white. The terms “bright box” and “grey box” attempt to capture the reality that you might still buy boxes from brand named vendors, but rather than buying them for their software capabilities, you can now buy them for their hardware capabilities.

The post On the ‘web: Hyperconvergence in the Network appeared first on rule 11 reader.

Network slicing will play key role in 5G networks

Single wireless networks will be separated into many, layered virtual networks when 5G is launched, experts say. The technique, called network slicing, is of a similar concept to software-defined networking (SDN), found now in some fixed networking, where managers program network behavior.The way dynamic network slicing will work is that communications specific to a particular 5G application, such as those found in Internet of Things (IoT) sensors or video, will be layered over the top of a common infrastructure, then software will manage the different service types.To read this article in full or to leave a comment, please click here

Worth Reading: Speeding up doesn’t help

The evolution then began, first at 1.5x speed, and eventually working up to twice the speed. At first, it made me laugh. Those distinct voices I had grown so accustomed to had turned into high-speed ramblings. Listening to Ira Glass and Terry Gross at 2x speed is like expecting a massage at a spa and getting a defibrillator instead. But you get used to it, and plus: 50-minute podcasts were taking me 25 minutes to get through. Those that were an hour or longer were flying by. I was no longer just getting through my queue; I now had time to discover even more podcasts. —Paolo Uggetti @ The Ringer

The post Worth Reading: Speeding up doesn’t help appeared first on rule 11 reader.

Tomorrow is the big day!

How fast time flies! Tomorrow ( August 10) is my JNCIE-DC lab day.

I spent last couple of days repeating my notes and labbing small optional topics like CoPP, ZTP, etc; and also familiar ones like CoS and MPLS L3VPN.

Today is the rest day. Fly to Amsterdam, drink a couple of beers and go to bed. Fortunately I’ve been there before, so no worries about how to find Juniper office and be late for the exam.

Plan for tomorrow: go to the lab and get the job done.

Be a Budget Hero with Docker Enterprise Edition

We recently started a multi-part learning series for SysAdmins and IT professionals called IT Starts With Docker. We started with the basics, covering container technology and Docker solutions for the enterprise. Now, we shift to the important question: Is it worth your time and your company’s investment to further explore Docker Enterprise Edition (EE)?

The resounding answer to that question is YES. IT teams who have adopted Docker EE are finding it faster and easier to deploy and maintain their applications, plus drive better infrastructure utilization, all without touching the underlying code. Developer teams are realizing productivity gains of their own by being able to onboard new developers faster, shortening the cycles from development to production, and elimination of the burdensome “it worked on my machine” problems.

Docker EE

Try the simple ROI calculator for yourself. It takes just a couple of minutes and allows you to estimate your own savings with Docker EE and how you can become the budget hero of your department. Then, register for our live webinar on Tuesday, August 15th, The Business Value of Docker, where we will outline how organizations like your own are saving 50% on their total costs with Docker EE. Continue reading

Episode 9 – What Your Mamma Never Told You About OSPF

As a continuation of our deep dives into routing protocols, this episode of Network Collective takes a closer look at OSPF. What does graph theory, CAP theorem, and distributed databases have to do with routing? What exactly is that type 4 LSA for? What is OSPF really good for? These and other questions are discussed by our guests Russ White and Nick Russo.

Russ White
Guest
Nicholas Russo
Guest
Jordan Martin
Co-Host
Eyvonne Sharp
Co-Host

Outro Music:
Danger Storm Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 3.0 License
http://creativecommons.org/licenses/by/3.0/

The post Episode 9 – What Your Mamma Never Told You About OSPF appeared first on Network Collective.

1 … 1,962 1,963 1,964 1,965 1,966 … 3,852