MQSim is a fast and accurate simulator modeling the performance of modern multi-queue (MQ) SSDs as well as traditional SATA based SSDs. MQSim faithfully models new high-bandwidth protocol implementations, steady-state SSD conditions, and the full end-to-end latency of requests in modern SSDs. It is described in detail in the FAST 2018 paper by Arash Tavakkol et al., "MQSim: A Framework for Enabling Realistic Studies of Modern Multi-Queue SSD Devices" (https://people.inf.ethz.ch/omutlu/pub/MQSim-SSD-simulation-framework_fast18.pdf)
I want to reproduce FAST '18 paper (especially the response time CDF and the time series plot of read/write response time in Appendix 3 and 4 respectively).
I cannot find out how to extract the response time of each request. The simulator's output only provides the average/min/max latencies/bandwidth of read/write requests.
I quickly checked the code (i.e., NVMe/SATA_consume_io_request functions), but the simulator doesn't store the response time (device_response_time) of each request. It just compares the response time with minimum and maximum values to keep track of min/max.
Is there any way to obtain the latency results of each request?
I missed configurations. I can get the response time log by setting "Enable_ResponseTime_Logging" and "ResponseTime_Logging_Period_Length" in Host_Parameter_Set.
joonsung-kim
commented
5 years ago
Hi. I am Joonsung Kim.
I want to reproduce FAST '18 paper (especially the response time CDF and the time series plot of read/write response time in Appendix 3 and 4 respectively). I cannot find out how to extract the response time of each request. The simulator's output only provides the average/min/max latencies/bandwidth of read/write requests. I quickly checked the code (i.e., NVMe/SATA_consume_io_request functions), but the simulator doesn't store the response time (device_response_time) of each request. It just compares the response time with minimum and maximum values to keep track of min/max.
Is there any way to obtain the latency results of each request?
Great thanks.