| 发明名称 | Acceleration prediction in hybrid systems | ||
| 摘要 | Predicting acceleration in a hybrid system may comprise determining a number of cross system calls in a first host-accelerator computer architecture running a workload. Host machine overhead and accelerator overhead in the first host-accelerator computer architecture associated with each of the cross system calls may be determined. Communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload may be determined. An application response time may be predicted for a candidate application to be run in a second host-accelerator computer architecture, based at least on the determined host machine overhead, the accelerator overhead, and the communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload. | ||
| 申请公布号 | US9348664(B2) | 申请公布日期 | 2016.05.24 |
| 申请号 | US201514718607 | 申请日期 | 2015.05.21 |
| 申请人 | International Business Machines Corporation | 发明人 | Dube Parijat;Meng Xiaoqiao;Zhang Li |
| 分类号 | G06F9/54;G06F9/445;G06F9/48;G06F9/50 | 主分类号 | G06F9/54 |
| 代理机构 | Scully, Scott, Murphy & Presser, P.C. | 代理人 | Scully, Scott, Murphy & Presser, P.C. ;Garg, Esq. Nidhi |
| 主权项 | 1. A method for predicting acceleration in a hybrid system, comprising: determining a number of cross system calls in a first host-accelerator computer architecture running a workload; determining host machine overhead and accelerator overhead in the first host-accelerator computer architecture associated with each of the cross system calls; determining communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload; and predicting an application response time for a candidate application to be run in a second host-accelerator computer architecture, based at least on the determined host machine overhead, the accelerator overhead, and the communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload, wherein the determining communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload comprises at least: determining communication delay associated with each of the cross system calls in the first host-accelerator computer architecture running a workload, based on parameter inferencing using response time and central processing unit (CPU) utilization measurement of the first host-accelerator computer architecture running the workload. | ||
| 地址 | Armonk NY US | ||