Begin forwarded message: From: "Chinthamani, Sundaram" <[email protected]<mailto:[email protected]>> Subject: RE: Stream Date: April 27, 2016 at 10:58:11 AM CDT To: "Sodani, Avinash" <[email protected]<mailto:[email protected]>>, "Kumaran, Kalyan" <[email protected]<mailto:[email protected]>> Cc: "Allison, CM" <[email protected]<mailto:[email protected]>>, "Kirkendall, Keith G" <[email protected]<mailto:[email protected]>>, "Fromkin, Russ" <[email protected]<mailto:[email protected]>>, "Chinthamani, Sundaram" <[email protected]<mailto:[email protected]>> Thanks. Then, All2All (420 GB/s) and Quad mode (440 GB/s) are delivering performance as expected. At this point, we are checking on whether we can re-produce this issue internally? When I get time, I will also log-in into your system and see what is going on? -Thanks, -Sundaram From: Sodani, Avinash Sent: Wednesday, April 27, 2016 8:32 AM To: Kumaran, Kalyan <[email protected]<mailto:[email protected]>>; Chinthamani, Sundaram <[email protected]<mailto:[email protected]>> Cc: Allison, CM <[email protected]<mailto:[email protected]>>; Kirkendall, Keith G <[email protected]<mailto:[email protected]>>; Fromkin, Russ <[email protected]<mailto:[email protected]>>; Sodani, Avinash <[email protected]<mailto:[email protected]>> Subject: RE: Stream Thanks Kumar. That at least confirms that quadrant mode bandwidth is as expected. -Avinash From: Kumaran, Kalyan [mailto:[email protected]] Sent: Wednesday, April 27, 2016 7:24 AM To: Sodani, Avinash; Chinthamani, Sundaram Cc: Allison, CM; Kirkendall, Keith G; Fromkin, Russ Subject: Stream Begin forwarded message: From: Scott Parker <[email protected]<mailto:[email protected]>> Subject: Re: [JLSE-Intel-NDA] Intel Meeting Notes Date: April 26, 2016 at 5:50:32 PM CDT Re-ran stream in Quadrant mode and I was able to get 435 GB/s with 1 thread per core, and a very slightly better 440 GB/s with 2 threads per core. In SNC-4 mode I saw 311 GB/s with 1 thread and 442 GB/s with 2 threads. All2All hit 425 GB/s with 1 thread per core. A peak of around 440 GB/s is basically what we should be expecting to see for a Bin 3 part, so the MCDRAM seems to be capable of delivering the expected bandwidth. However, SNC-4 is an outlier in requiring 128 threads, and gives unexpected low (73 GB/s) bandwidth from one of the clusters (vs 112 GB/s from each of the other three). -Scott