Number of requests per batch
Max parallel requests
Randomness in request duration
For this configuration, batched execution often provides a good balance between parallelism and resource usage.
Fastest Method
Batched
76.6% faster than slowest
Total Time
912ms
Executes one request at a time, waiting for each to complete before starting the next
Throughput
5.1 req/s
Avg Latency
195ms
Min Latency
160ms
Max Latency
239ms
Executes multiple requests simultaneously, limited by concurrency setting
Throughput
20.1 req/s
Avg Latency
193ms
Min Latency
162ms
Max Latency
239ms
Divides requests into batches, executing each batch in parallel, then moving to the next batch
Throughput
21.9 req/s
Avg Latency
206ms
Min Latency
160ms
Max Latency
233ms
馃弳 Winner!
76.6% faster than the slowest method