# Why is the product of average latency and average throughput greater than the number of clients (Little's Law)?

**URL:** <https://discuss.elastic.co/t/why-is-the-product-of-average-latency-and-average-throughput-greater-than-the-number-of-clients-littles-law/357811>\
**Category:** Elasticsearch\
**Tags:** rally\
**Created:** [April 19, 2024, 4:17pm UTC](https://discuss.elastic.co/t/why-is-the-product-of-average-latency-and-average-throughput-greater-than-the-number-of-clients-littles-law/357811 "2024-04-19T16:17:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![icyclv](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/icyclv/32/133680_2.png) [@icyclv](https://discuss.elastic.co/u/icyclv)\
**Post date:** [April 19, 2024, 4:17pm UTC](https://discuss.elastic.co/t/why-is-the-product-of-average-latency-and-average-throughput-greater-than-the-number-of-clients-littles-law/357811/1 "2024-04-19T16:17:25Z")

</div>

Hello,  
I have recently studied Little's law, and I want to verify my understanding through the data from Rally, envisioning that the number of clients = average latency x average throughput.  
I have selected the track and challenge as follows:  
--track=geonames --challenge=append-fast-with-conflicts  
But when querying rally-metrics, it seems that the result is greater than the number of clients (8):

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/4/2/42d4a041647db3d581934992abfb8e487eed2c28.png)

The average throughput is 60754 docs/s, divided by the Bulk size (5000) to get 12.15 ops/s, and the average latency is 0.703s, so their product is 8.54.

Is there anything I might have a problem with, or any details I might have overlooked? 😯I am quite puzzled by these results, hoping for your help to solve my problem, thank you very much.