# Aerospike read write latency metrics for namespace

**URL:** <https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647>\
**Category:** Monitoring\
**Created:** [November 5, 2018, 6:04pm UTC](https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647 "2018-11-05T18:04:38Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ren](https://avatars.discourse-cdn.com/v4/letter/r/ed655f/32.png) [@Ren](https://discuss.aerospike.com/u/Ren)\
**Post date:** [November 5, 2018, 6:04pm UTC](https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647/1 "2018-11-05T18:04:38Z")

</div>

Hi, I’m looking for read/write latency metrics for each namespace for overall cluster (not node level), using asloglatency i can get the latency for each namespace on each node but is there a way to pull these metrics to have graph for history for each namespace.

---

<div class="post-metadata">

**Author:** ![Albot](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/albot/32/2076_2.png) [@Albot](https://discuss.aerospike.com/u/Albot)\
**Post date:** [November 5, 2018, 8:55pm UTC](https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647/2 "2018-11-05T20:55:44Z")

</div>

You can go into asadm and run `show latency` to get a response from all nodes - but it wont give you a graph. If you want to graph this data out you’ll need to put it somewhere else. The best way is probably to setup a monitoring system that you can use to graph or use as a data source for a graphing system. [https://www.aerospike.com/docs/operations/monitor/](https://www.aerospike.com/docs/operations/monitor/)

---

<div class="post-metadata">

**Author:** ![Ren](https://avatars.discourse-cdn.com/v4/letter/r/ed655f/32.png) [@Ren](https://discuss.aerospike.com/u/Ren)\
**Post date:** [November 5, 2018, 9:46pm UTC](https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647/3 "2018-11-05T21:46:11Z")

</div>

but show latency only gives the latencies for node level. but I’m looking for namespace level latencies for overall cluster

---

<div class="post-metadata">

**Author:** ![Albot](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/albot/32/2076_2.png) [@Albot](https://discuss.aerospike.com/u/Albot)\
**Post date:** [November 5, 2018, 9:53pm UTC](https://discuss.aerospike.com/t/aerospike-read-write-latency-metrics-for-namespace/5647/4 "2018-11-05T21:53:39Z")

</div>

Ah, you’re right. In that case, inside asadm you can run `asinfo -v 'latency:' -l`. If you want to filter this further you could save the output somewhere and do more processing `asadm -e 'asinfo -v "latency:" -l'` - like this:

```auto
$ asadm -e 'asinfo -v "latency:" -l'|grep -A1 'mynamespace'|head
{mynamespace}-read:21:54:23-GMT,ops/sec,>1ms,>8ms,>64ms
21:54:33,19018.4,0.00,0.00,0.00
{mynamespace}-write:21:54:23-GMT,ops/sec,>1ms,>8ms,>64ms
21:54:33,111.5,0.00,0.00,0.00
--
{mynamespace}-read:21:54:25-GMT,ops/sec,>1ms,>8ms,>64ms
21:54:35,18829.8,0.00,0.00,0.00
{mynamespace}-write:21:54:25-GMT,ops/sec,>1ms,>8ms,>64ms
21:54:35,104.5,0.00,0.00,0.00
--

```