Andy Patrizio

Author Archives: Andy Patrizio

Google’s Partner Interconnect connects SMBs to its data centers

If you are a large-scale enterprise, Google has a service called Dedicated Interconnect that offers 10Gbps connections between your data center and one of theirs. But what if you are a smaller firm and don’t need that kind of bandwidth and the expense that goes with it?Google now has you covered. The cloud giant recently announced Google Cloud Partner Interconnect, a means of establishing a direct connection between a SMB data center, with emphasis on the medium-sized business, and Google's hybrid cloud platform. The company did this in concert with 23 ISP partners around the globe.To read this article in full, please click here

Google’s Partner Interconnect connects SMBs to its data centers

If you are a large-scale enterprise, Google has a service called Dedicated Interconnect that offers 10Gbps connections between your data center and one of theirs. But what if you are a smaller firm and don’t need that kind of bandwidth and the expense that goes with it?Google now has you covered. The cloud giant recently announced Google Cloud Partner Interconnect, a means of establishing a direct connection between a SMB data center, with emphasis on the medium-sized business, and Google's hybrid cloud platform. The company did this in concert with 23 ISP partners around the globe.To read this article in full, please click here

Cyxtera offers on-demand data center provisioning software

Cyxtera Technologies has launched the Cyxtera Extensible Data Center (CXD) platform, a software platform for data centers that offers customers rapid on-demand provisioning to a host of colocation and connectivity services.Through a combination of a network and services provisioning engine and an intra-data center software-defined network fabric, the CXD platform allows colocation customers to provision services on demand or via a web console.CXD brings cloud-like experience to colocation CXD comes with two key features: the Unified Services Port and Network Exchange. The Unified Services Port enables access to multiple data center services over a single physical port, while the Network Exchange provides automated provisioning to select network service providers. The caveat is they must also be running CXD.To read this article in full, please click here

Cyxtera offers on-demand data center provisioning software

Cyxtera Technologies has launched the Cyxtera Extensible Data Center (CXD) platform, a software platform for data centers that offers customers rapid on-demand provisioning to a host of colocation and connectivity services.Through a combination of a network and services provisioning engine and an intra-data center software-defined network fabric, the CXD platform allows colocation customers to provision services on demand or via a web console.CXD brings cloud-like experience to colocation CXD comes with two key features: the Unified Services Port and Network Exchange. The Unified Services Port enables access to multiple data center services over a single physical port, while the Network Exchange provides automated provisioning to select network service providers. The caveat is they must also be running CXD.To read this article in full, please click here

Is Facebook looking to build its own data center chips?

A job posting on Facebook has led to speculation that the company is building a team to design its own semiconductors, thus ending their reliance on Intel. If so, it would be another step in the trend of major firms building their own silicon.Bloomberg was the first to note a job opening, titled “Manager, ASIC Development,” that sought a manager to help build an "end-to-end SoC/ASIC, firmware and driver development organization." There is also an opening for an “ASIC & FPGA Design Engineer,” which seems an unusual position for a social network website to need.To read this article in full, please click here

Is Facebook looking to build its own data center chips?

A job posting on Facebook has led to speculation that the company is building a team to design its own semiconductors, thus ending their reliance on Intel. If so, it would be another step in the trend of major firms building their own silicon.Bloomberg was the first to note a job opening, titled “Manager, ASIC Development,” that sought a manager to help build an "end-to-end SoC/ASIC, firmware and driver development organization." There is also an opening for an “ASIC & FPGA Design Engineer,” which seems an unusual position for a social network website to need.To read this article in full, please click here

Cray reunites with AMD for new supercomputers

Cray owes its survival to AMD. The company was bought by SGI in 1996, hollowed out, and spun off in 2000 with very little left. SGI had taken most of the talent and IP.Desperate for a win, Cray began working with Sandia National Labs in 2002 to build a supercomputer based on x86 technology. Intel at the time was dismissive of 64-bit x86 and was promoting Itanium. AMD had other plans and was in the process of developing Athlon for desktops and Opteron for servers.[ Learn how server disaggregation can boost data center efficiency and find out what the top 10 fastest supercomputers are. | Get regularly scheduled insights by signing up for Network World newsletters. ] The project came to be known as Red Storm, starting with single-core Opterons and upgrading to dual- and quad-core CPUs as they hit the market. Red Storm ranked as high as number two on the Top 500 list of supercomputers. More important, it served as the basis for the XT3 line of supercomputers that revived Cray as a player in that field, and lit a fire under Intel as well.To read this article in full, please click here

Cray reunites with AMD for new supercomputers

Cray owes its survival to AMD. The company was bought by SGI in 1996, hollowed out, and spun off in 2000 with very little left. SGI had taken most of the talent and IP.Desperate for a win, Cray began working with Sandia National Labs in 2002 to build a supercomputer based on x86 technology. Intel at the time was dismissive of 64-bit x86 and was promoting Itanium. AMD had other plans and was in the process of developing Athlon for desktops and Opteron for servers.[ Learn how server disaggregation can boost data center efficiency and find out what the top 10 fastest supercomputers are. | Get regularly scheduled insights by signing up for Network World newsletters. ] The project came to be known as Red Storm, starting with single-core Opterons and upgrading to dual- and quad-core CPUs as they hit the market. Red Storm ranked as high as number two on the Top 500 list of supercomputers. More important, it served as the basis for the XT3 line of supercomputers that revived Cray as a player in that field, and lit a fire under Intel as well.To read this article in full, please click here

New marketplace for FPGA custom apps launches

A French company called Accelize has launched AccelStore, an app store specifically around providing custom programmed applications for FPGA accelerators.FPGAs are dedicated processors known for doing two things: very fast processing, and being reprogrammable. CPUs have to be general-purpose processors that run an OS, but an FPGA has the luxury of doing a dedicated task, so the architecture is different.The problem is that while FPGAs are reprogrammable to do new, specific tasks, they aren’t that easy to program. In fact, it’s often pretty hard to do. That’s Accelize’s sales pitch. Rather than writing the code to reprogram the FPGAs in your servers, it has the templates for you.To read this article in full, please click here

New marketplace for FPGA custom apps launches

A French company called Accelize has launched AccelStore, an app store specifically around providing custom programmed applications for FPGA accelerators.FPGAs are dedicated processors known for doing two things: very fast processing, and being reprogrammable. CPUs have to be general-purpose processors that run an OS, but an FPGA has the luxury of doing a dedicated task, so the architecture is different.The problem is that while FPGAs are reprogrammable to do new, specific tasks, they aren’t that easy to program. In fact, it’s often pretty hard to do. That’s Accelize’s sales pitch. Rather than writing the code to reprogram the FPGAs in your servers, it has the templates for you.To read this article in full, please click here

One in five serverless apps has a critical security vulnerability

Serverless computing is an emerging trend that is likely to explode in popularity this year. It takes the idea of a smaller server footprint to the next level. First, there were virtual machines, which ran a whole instance of an operating system. Then they were shrunk to containers, which only loaded the bare minimum of the OS required to run the app. This led to a smaller footprint.Now we have “serverless” apps, which is a bit of a misnomer. They still run on a server; they just don’t have a dedicated server, virtual machine, or container running 24/7. They run in a server instance until they complete their task, then shut down. It’s the ultimate in small server footprint and reducing server load.To read this article in full, please click here

One in five serverless apps has a critical security vulnerability

Serverless computing is an emerging trend that is likely to explode in popularity this year. It takes the idea of a smaller server footprint to the next level. First, there were virtual machines, which ran a whole instance of an operating system. Then they were shrunk to containers, which only loaded the bare minimum of the OS required to run the app. This led to a smaller footprint.Now we have “serverless” apps, which is a bit of a misnomer. They still run on a server; they just don’t have a dedicated server, virtual machine, or container running 24/7. They run in a server instance until they complete their task, then shut down. It’s the ultimate in small server footprint and reducing server load.To read this article in full, please click here

One in five serverless apps has a critical security vulnerability

Serverless computing is an emerging trend that is likely to explode in popularity this year. It takes the idea of a smaller server footprint to the next level. First, there were virtual machines, which ran a whole instance of an operating system. Then they were shrunk to containers, which only loaded the bare minimum of the OS required to run the app. This led to a smaller footprint.Now we have “serverless” apps, which is a bit of a misnomer. They still run on a server; they just don’t have a dedicated server, virtual machine, or container running 24/7. They run in a server instance until they complete their task, then shut down. It’s the ultimate in small server footprint and reducing server load.To read this article in full, please click here

Overclock puts your idle servers to work for other people

Putting unused CPUs to work is nothing new. In the modern era, it started in 1999 when the SETI Institute launched [email protected], a screensaver that also examined slices of radio signals gathered by a giant telescope for signs of intergalactic life. Nineteen years later, and ET still hasn’t phoned us.But the concept grew to dozens of science and math-related projects. I took part in the World Community Grid run by IBM for years, letting my idle PC look for potential cures for AIDS and Ebola.To read this article in full, please click here

Overclock puts your idle servers to work for other people

Putting unused CPUs to work is nothing new. In the modern era, it started in 1999 when the SETI Institute launched [email protected], a screensaver that also examined slices of radio signals gathered by a giant telescope for signs of intergalactic life. Nineteen years later, and ET still hasn’t phoned us.But the concept grew to dozens of science and math-related projects. I took part in the World Community Grid run by IBM for years, letting my idle PC look for potential cures for AIDS and Ebola.To read this article in full, please click here

New distributed database adds international and GDPR controls

The European Union’s General Data Protection Regulation (GDPR) will force very strict new privacy compliance rules on firms doing business in the EU, but a startup that has an atrocious company and product name has what it says is the solution to maintaining compliance.Cockroach Labs has introduced version 2.0 of its CockroachDB distributed database, which can be run in a data center or cloud. The company bills the product as “the SQL database for global cloud services.” It automatically scales, rebalances, and repairs databases spread over multiple locations.To read this article in full, please click here

New distributed database adds international and GDPR controls

The European Union’s General Data Protection Regulation (GDPR) will force very strict new privacy compliance rules on firms doing business in the EU, but a startup that has an atrocious company and product name has what it says is the solution to maintaining compliance.Cockroach Labs has introduced version 2.0 of its CockroachDB distributed database, which can be run in a data center or cloud. The company bills the product as “the SQL database for global cloud services.” It automatically scales, rebalances, and repairs databases spread over multiple locations.To read this article in full, please click here

Nvidia packs 2 petaflops of performance in a single compact server

At its GPU Technology Conference this week, Nvidia took the wraps off a new DGX-2 system it claims is the first to offer multi-petaflop performance in a single server, thus greatly reducing the footprint to get to true high-performance computing (HPC).DGX-2 comes just seven months after the DGX-1 was introduced, although it won’t ship until the third quarter. However, Nvidia claims it has 10 times the compute power as the previous generation thanks to twice the number of GPUs, much more memory per GPU, faster memory, and a faster GPU interconnect.[ Learn how server disaggregation can boost data center efficiency. | Get regularly scheduled insights by signing up for Network World newsletters. ] The DGX-2 uses a Tesla V100 CPU, the top of the line for Nvidia’s HPC and artificial intelligence-based cards. With the DGX-2, it has doubled the on-board memory to 32GB. Nvidia claims the DGX-2 is the world’s first single physical server with enough computing power to deliver two petaflops, a level of performance usually delivered by hundreds of servers networked into clusters.To read this article in full, please click here

Nvidia packs 2 petaflops of performance in a single compact server

At its GPU Technology Conference this week, Nvidia took the wraps off a new DGX-2 system it claims is the first to offer multi-petaflop performance in a single server, thus greatly reducing the footprint to get to true high-performance computing (HPC).DGX-2 comes just seven months after the DGX-1 was introduced, although it won’t ship until the third quarter. However, Nvidia claims it has 10 times the compute power as the previous generation thanks to twice the number of GPUs, much more memory per GPU, faster memory, and a faster GPU interconnect.[ Learn how server disaggregation can boost data center efficiency. | Get regularly scheduled insights by signing up for Network World newsletters. ] The DGX-2 uses a Tesla V100 CPU, the top of the line for Nvidia’s HPC and artificial intelligence-based cards. With the DGX-2, it has doubled the on-board memory to 32GB. Nvidia claims the DGX-2 is the world’s first single physical server with enough computing power to deliver two petaflops, a level of performance usually delivered by hundreds of servers networked into clusters.To read this article in full, please click here

Patches for Meltdown and Spectre aren’t that bad after all

Internal tests from a leading industry vendor have shown that fixes applied to servers running Linux or Windows Server aren’t as detrimental as initially thought, with many use cases seeing no impact at all.The Meltdown and Spectre vulnerabilities, first documented in January, seemed like a nightmare for virtualized systems, but that is overblown. There are a lot of qualifiers, starting with what you are doing and what generation processor you are using.The tests were done on servers running Xeons of the Haswell-EP (released in 2014), Broadwell-EP (released in 2016), and Skylake-EP (released in 2017). Haswell and Broadwell were the same microarchitecture, with minor tweaks. The big change there was Broadwell was a die shrink. Skylake, though, was a whole new architecture, and as it turns out, that made the difference.To read this article in full, please click here