Remember how hard it was to get your hands on an ARM server? Well, those days may very well be behind us as five computer makers have announced servers with ARM processors. These new servers are direct competition to x86 systems in the mainstream market and are primarily designed for internet and cloud workloads. In addition to that, they also have the 48-core Cavium ThunderX chip, which is based on 64-bit ARM architecture.
The five companies making these servers are Gigabyte, Inventec, Wistron, Penguin Computing, and E4 Computer Engineering and these servers will be based on designs that are popular in x86 servers though they will have ARM processors. However, there is an even more interesting aspect to this and that is that some of the new servers will have the ability to use Nvidia's Tesla graphics processors, which adds extra processing power for graphics, engineering, and other computing applications that require high-performance.
All of the systems generally have one or two sockets and all of them have different strengths. Gigabyte's servers can be configured with up to 24 2.5" hard drives, which makes it perfect for web serving or storage. Penguin Computing's 19" Valkre system will be shipping in 2016 and is aimed at high-performance computing and has the ability to be configured with SSDs and different I/O technologies. Wistron's WV-S7224-10 and WV-A7424 are 2U and 4U storage servers. All of the servers, however, share power and cooling resources.
A majority of the companies announcing ARM servers are capable of making and supplying servers to buyers directly, which eliminates the middleman that is usually involved in the whole selling process. Companies like Wistron and Inventec are also capable of making an impact as server vendors in China whereas some of the other companies cannot. Pricing for the servers has yet to become available from any of the five companies though the servers themselves were announced at the Supercomputing 15 conference in Austin, Texas this past week.
ARM develops and licenses its processor architecture and is best known for its mobile chips found in most smartphones. A lot of people believe that the power efficiency that is derived from the mobile chips could very well translate to low-power ARM servers though as of now only a couple of systems have been made available. The most noteworthy system is the Moonshot from Hewlett-Packard Enterprise.
These new servers could be the motivation that ARM needs to make a serious push towards dominating the server market. Chip makers are having a very hard time generating profits with mobile and PC chips considering margins are thin for those systems. As a result, servers could be a very profitable alternative for vendors using ARM-architecture chips. AMD is planning on offering both x86-architecture and ARM-architecture chips as it continues to rebuild its server product line.
Content originally published here
Sharing this story on social media? Use these hashtags! #Servers #Intel #ARM #AMD
Friday, November 20, 2015
Looking For An Alternative To Intel? ARM Chips May Be That Answer
Labels:
AMD,
ARM,
Intel,
server chips,
Servers
Monday, November 16, 2015
Oracle Made Huge Server Announcement At OpenWorld
Five years ago, when Oracle bought Sun Microsystems, Larry Ellison talked a lot about how owning the entire systems stack would allow him to do unique things with the servers from Oracle. After five years of hearing the hype and the talk, Ellison looks like he may have finally delivered. However, the big question still remains whether or not consumers will actually buy what he's selling.
Oracle has announced a brand new line of servers at OpenWorld recently that are based on a new Sparc processor known as the M7. This upgrade has the typical improvements that you would expect in a new chip, including more cores, bigger caches, and higher bandwidth though the most interesting thing is software functions that Oracle has embedded into the silicon that improves the performance and security of applications. These include a memory-protection technology that is capable of providing a new level of security for in-memory databases along with an acceleration engine that allows data to be decompressed in near-real time for analytics. As a result, you get a wider use of compressed data.
According to Principal Analyst at Insight64 Nathan Brookwood, "Both of those are very interesting, because they're features I don't think a company that makes just chips -- that didn't have the software guys working with them -- would have invented." Brookwood has asked Oracle what new features it was able to include on each new processor the company has released as a result of owning both companies." Invariably they would say, well, you know, it takes time to do that, we don't really have anything yet," Brookwood added. "But with the M7, they do."
This is also the first new Sparc processor core designed entirely in-house by Oracle. It takes four to six years to design a new microprocessor, which has been the time frame that Oracle has owned Sun. "This is the first project that has Larry's fingerprints all over it," added Marshall Choy, Oracle Senior Director for Optimized Solutions. The M7 is currently on sale now and is in new models of Oracle's T-Series and M-Series servers along with an upgrade to the Oracle Supercluster, which is a pre-configured system for running the Oracle database.
The memory-protection technology, which is also known as "silicon-secured memory", prevents malicious programs from accessing parts of main memory that they're not used to, thus disposing of a common attack method for hackers. Whenever a new application needs new memory, the M7 creates a unique "color bit" which ensures the application can access only the portion of memory assigned to it. When the application process ends, the color bit expires and a new one is created for the next allocation of memory. "That's how we can prevent a piece of malware from accessing a memory segment it's not authorized to, because it will do that color code checking and abort the program if it doesn't match," Choy added.
This is particularly significant due to the fact that customers are storing large amounts of data in memory for analytics, where it is more susceptible to attacks. The secured memory technology will be available to any application that runs on the M7 systems, Choy said, and not just ones from Oracle. In addition to that, it is also capable of uncovering low-level bugs in software because it exposes any problems with memory allocation. For decompression, the accelerator in the chip will run at the full speed of Oracle's in-memory database. This means that customers will be able to use compressed data for in-memory computing without the performance overhead that they would normally have to deal with.
Oracle is offering the M7 chip in the T-Series servers, which are typically used for scale-out configurations, and in the M-Series servers, which scale up to form big SMP boxes. This is also the first time the company will use the same processor across both product lines and, according to Choy, "We literally have one chip. We have exactly on part number for the M7."
Content originally published here
Sharing this story on Social Media? Use these hashtags! #Oracle #OpenWorld #M7 #Servers
Oracle has announced a brand new line of servers at OpenWorld recently that are based on a new Sparc processor known as the M7. This upgrade has the typical improvements that you would expect in a new chip, including more cores, bigger caches, and higher bandwidth though the most interesting thing is software functions that Oracle has embedded into the silicon that improves the performance and security of applications. These include a memory-protection technology that is capable of providing a new level of security for in-memory databases along with an acceleration engine that allows data to be decompressed in near-real time for analytics. As a result, you get a wider use of compressed data.
According to Principal Analyst at Insight64 Nathan Brookwood, "Both of those are very interesting, because they're features I don't think a company that makes just chips -- that didn't have the software guys working with them -- would have invented." Brookwood has asked Oracle what new features it was able to include on each new processor the company has released as a result of owning both companies." Invariably they would say, well, you know, it takes time to do that, we don't really have anything yet," Brookwood added. "But with the M7, they do."
This is also the first new Sparc processor core designed entirely in-house by Oracle. It takes four to six years to design a new microprocessor, which has been the time frame that Oracle has owned Sun. "This is the first project that has Larry's fingerprints all over it," added Marshall Choy, Oracle Senior Director for Optimized Solutions. The M7 is currently on sale now and is in new models of Oracle's T-Series and M-Series servers along with an upgrade to the Oracle Supercluster, which is a pre-configured system for running the Oracle database.
The memory-protection technology, which is also known as "silicon-secured memory", prevents malicious programs from accessing parts of main memory that they're not used to, thus disposing of a common attack method for hackers. Whenever a new application needs new memory, the M7 creates a unique "color bit" which ensures the application can access only the portion of memory assigned to it. When the application process ends, the color bit expires and a new one is created for the next allocation of memory. "That's how we can prevent a piece of malware from accessing a memory segment it's not authorized to, because it will do that color code checking and abort the program if it doesn't match," Choy added.
This is particularly significant due to the fact that customers are storing large amounts of data in memory for analytics, where it is more susceptible to attacks. The secured memory technology will be available to any application that runs on the M7 systems, Choy said, and not just ones from Oracle. In addition to that, it is also capable of uncovering low-level bugs in software because it exposes any problems with memory allocation. For decompression, the accelerator in the chip will run at the full speed of Oracle's in-memory database. This means that customers will be able to use compressed data for in-memory computing without the performance overhead that they would normally have to deal with.
Oracle is offering the M7 chip in the T-Series servers, which are typically used for scale-out configurations, and in the M-Series servers, which scale up to form big SMP boxes. This is also the first time the company will use the same processor across both product lines and, according to Choy, "We literally have one chip. We have exactly on part number for the M7."
Content originally published here
Sharing this story on Social Media? Use these hashtags! #Oracle #OpenWorld #M7 #Servers
Labels:
M7,
OpenWorld,
Oracle,
Oracle M-Series,
Oracle T-Series,
server chips,
Servers,
Sun
Subscribe to:
Posts (Atom)