command ‘top’:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
32383 kudu 20 0 21.735g 1.641g 32404 S
176.0 1.3 714:52.34 kudu-tserver
command ‘sar -n DEV 1’:
Average: IFACE rxpck/s txpck/s rxkB/s txkB/s rxcmp/s txcmp/s rxmcst/s
Average: eth0 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Average: eth1 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Average: eth2 4735.46 185349.84 447.21
273590.82 0.00 0.00 0.00
Average: eth3 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Average: lo 59.84 59.84 108.97 108.97 0.00 0.00 0.00
command ‘top’:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
21684 kudu 20 0 18.456g 2.536g 32772 S
6.2 2.0 263:12.00 kudu-tserver
Command ‘sar -n DEV 1’:
10:31:52 AM IFACE rxpck/s txpck/s rxkB/s txkB/s rxcmp/s txcmp/s rxmcst/s
10:31:53 AM eth0 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:31:53 AM eth1 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:31:53 AM eth2 188.75 188.75 30.13
31.23 0.00 0.00 0.00
10:31:53 AM eth3 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:31:53 AM lo 18.75 18.75 10.71 10.71 0.00 0.00 0.00
Hi Helifu
Is it possible that you created this table initially with only one tablet, and then later tried the same job with the presplit option? I'm wondering if somehow your table has only one tablet. You can check the Kudu master web UI to see.
The other thing to note is that currently the "scan" workload is not well implemented in the YCSB client.
The Java client doesn't support a client-side LIMIT and so the scans all fetch significantly more data than expected. So, the scan performance measured by YCSB is not very accurate, and we haven't really tested it. It's possible there is some other bug which is causing the scan workload to always read from the same part of the table.
Does the issue occur with the other workloads like get/insert/load?
-Todd
好吧,我掉坑里了,希望后面的同学不要掉进去了!
相关阅读: