{"id":4342,"date":"2018-10-10T06:23:18","date_gmt":"2018-10-10T06:23:18","guid":{"rendered":"https:\/\/support.plunify.com\/en\/?p=4342"},"modified":"2021-06-03T02:21:21","modified_gmt":"2021-06-03T02:21:21","slug":"fpga-expansion-pack-1-2-it-is-all-about-runtime","status":"publish","type":"post","link":"https:\/\/support.plunify.com\/en\/2018\/10\/10\/fpga-expansion-pack-1-2-it-is-all-about-runtime\/","title":{"rendered":"FPGA Expansion Pack 1.2 – All about runtime!"},"content":{"rendered":"

We released FPGA Expansion Pack (FEP) 1.2 today, and it is all about improving the runtime and the user experience. Here is what you can expect in this release.<\/p>\n

No Regeneration of Out-Of-Context modules<\/h2>\n

In previous versions, we regenerated the IP targets of all OOC modules by default to ensure that the IP was up-to-date. We did that because\u00a0of compatibility issues with different versions, cached files with our design optimization product, InTime. However, from version 1.2 onwards we will not regenerate IP targets by default. If there is a need to regenerate the IP, please do so before submitting to the cloud.<\/p>\n

Impact on runtime:
\n\"\"\"\"<\/p>\n

No Generation of Bitstream<\/h2>\n

This began as something that was meant to be convenient for the user. However, in cases where the IP is licensed, the inability to generate the bitstream resulted in errors and a waste of time and money. So FEP will not create bitstreams in the cloud by default.<\/p>\n

\"\"<\/p>\n

Impact on runtime:
\n\"\"<\/p>\n

New high-performance servers - Z1d<\/h2>\n

We have enabled these servers on Plunify Cloud. These are the latest high performance servers from\u00a0AWS<\/a>.<\/p>\n

Amazon EC2 z1d instances deliver high single thread performance due to a custom Intel\u00ae Xeon\u00ae Scalable processor with a sustained all-core frequency of up to 4.0 GHz, the fastest of any cloud instance. z1d provides both high compute performance and high memory, which is ideal for electronic design automation (EDA), gaming, and certain relational database workloads with high per-core licensing costs.<\/p><\/blockquote>\n

After performing some simple benchmarking, we found that they are on average 10-16% faster than running on the C5 class of server. (A 4 hour\u00a0build will complete in about 3h 20min)<\/p>\n

Cost-wise, it is about 17 - 60% higher, depending on the class of servers you are comparing with. Do note that the Plunify Cloud price is an all-inclusive pricing that includes tool licenses, bandwidth and storage.<\/strong><\/span><\/p>\n\n\n\n\n\n\n\n\n
AWS Server Type<\/strong><\/td>\nServer Class<\/strong><\/td>\nvCPU<\/strong><\/td>\nECU<\/strong><\/td>\nRAM<\/strong><\/td>\nPlunify Cloud Price ($)<\/strong><\/td>\n<\/tr>\n
c5.2xlarge<\/td>\n6<\/td>\n8<\/td>\n34<\/td>\n16<\/td>\n0.78<\/td>\n<\/tr>\n
c5.4xlarge<\/td>\n7<\/td>\n16<\/td>\n68<\/td>\n32<\/td>\n1.19<\/td>\n<\/tr>\n
z1.large<\/td>\nz1<\/td>\n2<\/td>\n15<\/td>\n16<\/td>\n0.64<\/td>\n<\/tr>\n
z1.xlarge<\/td>\nz2<\/td>\n4<\/td>\n28<\/td>\n32<\/td>\n0.92<\/td>\n<\/tr>\n
z1.2xlarge<\/td>\nz3<\/td>\n8<\/td>\n53<\/td>\n64<\/td>\n1.46<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Impact on runtime:
\n\"\"\"\"\"\"<\/p>\n

How to get it<\/h2>\n

To try the new version, please go to this page<\/a>\u00a0to download the newest FEP package.<\/p>\n

Subscribe to Plunify Blog<\/h4>\n

<\/h2>\n
\n\t\t\t\t

Enter your email address and have the latest insights on FPGA, cloud and Machine Learning delivered straight to your inbox.<\/p>\n<\/div>

\n

\n \n \n \n \n\t\t\t\t\t\t