# Benchmarking tiledb read performance

**URL:** <https://forum.tiledb.com/t/benchmarking-tiledb-read-performance/278>\
**Category:** Uncategorized\
**Created:** [December 1, 2020, 6:39am UTC](https://forum.tiledb.com/t/benchmarking-tiledb-read-performance/278 "2020-12-01T06:39:45Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![qiuwei](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.tiledb.com/qiuwei/32/80_2.png) [@qiuwei](https://forum.tiledb.com/u/qiuwei)\
**Post date:** [December 10, 2020, 2:35pm UTC](https://forum.tiledb.com/t/benchmarking-tiledb-read-performance/278/3 "2020-12-10T14:35:00Z")

</div>

@stavros

Following your advice, I did more benchmarks focusing on the 1D dense array modeling.  
The results are as follows:

 ![image](https://canada1.discourse-cdn.com/flex035/uploads/tiledb/original/1X/4c985f6163b57976203ea2397d6944a2547889c2.png)

It seems that TileDB is still quite behind the parquet, especially for the writing.

If you are willing to investigate this case more closely, I can share the dataset I used to do the benchmark.

---

_[View the full topic](https://forum.tiledb.com/t/benchmarking-tiledb-read-performance/278)._