Five Reasons You Should Consider Adding Acceleration Tech to Your Apps NowMonday March 30, 2020
By Julie DiBene
Director, Marketing Communications
Seems the entire world is in a holding pattern right now but in high tech, we move forward as best we can, albeit while many of us are sheltering in place. Many of us working remotely, most of us are in a wait and see mode. But this is also the perfect time to review your tech strategy, to take a hard look at what technoloiges are coming down the road and what you can put into place now to improve your applications and ultiamtely, drive your business forward.
Five reasons you should consider adding acceleration tech now.
- Cost. Let’s face it, the first thing you probably look at when considering a new technology is the cost. The good news is that one MoSys BE-2 has the equivalent density to four QDRs and one BE-3 has the equivalent density to eight QDRs. That adds up to huge savings. In addition, your product’s PCB can use less layers to route the signal traces which translates into even more savings. There is also the inherent ease of routing 64 Signal traces for BE verses hundreds of traces for a QDR. Plus, the MoSys BE uses less power than the equivalent density of QDRs so less power means less power supply and less cooling , all saving money. Finally, the MoSys memory controller RTL is available which saves design time and shorter design cycles always mean more savings.
- Those Pesky Bottlenecks. As most of us know the mere act of just moving data between storage and a processor is reaching a breaking point as far as execution time is concerned. The technology fix for this has been to move the processor elements closer to or even into the memory without paying a cost penalty for replicating processing elements. Performance scaling has also become key to remaining competitive while addressing the growing requirements being placed on the network. At MoSys we have integrated RISC engine cores on a single die with a large block of memory. This places the processing engines right at the edge of the memory array (or in MoSys’s case, in the center of the die) which minimizes the memory latency and maximizes the random-access throughput. In essence, MoSys is offering a way to move the processing of algorithms to the memory. Our PHE (Programmable HyperSpeed Engine) is the latest semiconductor device offering from MoSys. Our PHE has 1Gb of high-speed memory with 32 multi-threaded RISC cores, all on a single chip. The 1Gb of memory is intended to be the primary repository of data and not a cache. The RISC cores also have access to 2Mb of scratch pad SRAM. This allows customers to “offload” a high frequency, repetitive task to the PHE.
- The Need for Speed. Access speed and bandwidth both are frequently limited by the traditional wide parallel Address, Data bus structures used with traditional memories. MoSys Bandwidth Engine products offer an interface that can support future speed and maintain the high reliability that the traditional parallel bus structures provided. To address the issue of speed, both today and in the future, it is believed that SerDes addresses this issue. Common speeds of 10Gbps up to 28Gbps are readily available on FPGAs and soon speeds of up to 112Gbps will be commonly available. Transactions between the MoSys Accelerator Engines (BE-2 and BE-3) all run at full SerDes rate allowing up to 2.5B transactions per Sec when talking with the BE-2 device and up to 5B transactions per second when interfacing with the BE-3 device.
- Scaling. When considering a new technology, you have to think about how it can scale. In today’s fast paced markets, customers need to keep up with the need for increasing performance and ensuring that their software investment can be Ported to a range of increasing Performance Scalable Hardware. The MoSys’ “Virtual Accelerator Engine” products facilitate the ability to provide a Software Platform to accelerate key functions in high speed FPGA applications. The first platform, “Packet Classification” address such markets as networking, data center, security and AI by inspecting data and determining the proper routing. The MoSys VAE uses the Cloud concepts of a common software API that will execute on several hardware platforms with the ability to scale performance up to 100x from a software only solution to an accelerated FPGA design using MoSys Accelerator Engine hardware running the MoSys Graph Memory Engine IP. Scalability from MoSys also comes from the ability of each implementation to take advantage of available compute and memory resources. Memory can range from DRAM adjacent to the CPU, to block memory or HBM on the FPGA or 1T-SRAM tightly coupled with 32 RISC engines in MoSys Accelerator Engine IC products. The software common API is common across different FPGA configurations with the common RTL (VAE IF), giving customers more scalable options than ever before.
- Performance. Ah yes, the unicorn of tech solutions, performance looms large and wide when weighing the benefits of various solutions. All MoSys Accelerator Engines have Embedded In-Memory functions of BURST and RWM which are designed to execute much faster in-memory, than can be executed in traditional memory. Finally, Programmable User functions can be embedded in the Programmable HyperSpeed Engine that has 32 RISC Cores which allows another level of performance acceleration.
So, there you have it. Five reasons you should consider when adding an acceleration technology, all designed to take your applications to the next level.
If you are looking for more technical information or need to discuss your technical challenges with an expert, we are happy to help. Email us and we will arrange to have one of our technical specialists speak with you. You can also sign up for updates. Finally, please follow us on social media so we can keep in touch.