Update performance to 0.4.76-nightly

Author: BohuTANG

website/datafuse/docs/overview/performance.md

@@ -7,25 +7,25 @@ title: Performance
     * **Memory SIMD-Vector processing performance only**
     * Dataset: 100,000,000,000 (100 Billion)
     * Hardware: AMD Ryzen 7 PRO 4750U, 8 CPU Cores, 16 Threads
-    * Rust: rustc 1.55.0-nightly (868c702d0 2021-06-30)
+    * Rust: rustc 1.56.0-nightly (e3b1c12be 2021-08-02)
     * Build with Link-time Optimization and Using CPU Specific Instructions
 
-| Query                                                        | DatafuseQuery (v0.4.48-nightly)                         |
+| Query                                                        | DatafuseQuery (v0.4.76-nightly)                         |
 | ------------------------------------------------------------ | --------------------------------------------------- |
-| SELECT avg(number) FROM numbers_mt(100000000000)             | 4.35 s.<br /> (22.97 billion rows/s., 183.91 GB/s.) |
-| SELECT sum(number) FROM numbers_mt(100000000000)             | 4.20 s.<br />(23.79 billion rows/s., 190.50 GB/s.)  |
-| SELECT min(number) FROM numbers_mt(100000000000)             | 4.92 s.<br />(20.31 billion rows/s., 162.64 GB/s.)  |
-| SELECT max(number) FROM numbers_mt(100000000000)             | 4.77 s.<br />(20.95 billion rows/s., 167.78 GB/s.)  |
-| SELECT count(number) FROM numbers_mt(100000000000)           | 2.91 s.<br />(34.33 billion rows/s., 274.90 GB/s.)  |
-| SELECT sum(number+number+number) FROM numbers_mt(100000000000) | 19.83 s.<br />(5.04 billion rows/s., 40.37 GB/s.)   |
-| SELECT sum(number) / count(number) FROM numbers_mt(100000000000) | 3.90 s.<br />(25.62 billion rows/s., 205.13 GB/s.)  |
-| SELECT sum(number) / count(number), max(number), min(number) FROM numbers_mt(100000000000) | 8.28 s.<br />(12.07 billion rows/s., 96.66 GB/s.)   |
-| SELECT number FROM numbers_mt(10000000000) ORDER BY number DESC LIMIT 100 | 4.80 s.<br />(2.08 billion rows/s., 16.67 GB/s.)    |
-| SELECT max(number), sum(number) FROM numbers_mt(1000000000) GROUP BY number % 3, number % 4, number % 5 | 6.31 s.<br />(158.49 million rows/s., 1.27 GB/s.) |
+| SELECT avg(number) FROM numbers_mt(100000000000)             | 3.712 s.<br /> (26.94 billion rows/s., 215.52 GB/s.)|
+| SELECT sum(number) FROM numbers_mt(100000000000)             | 3.669 s.<br /> (27.26 billion rows/s., 218.07 GB/s.)|
+| SELECT min(number) FROM numbers_mt(100000000000)             | 4.498 s.<br /> (22.23 billion rows/s., 177.85 GB/s.)|
+| SELECT max(number) FROM numbers_mt(100000000000)             | 4.438 s.<br /> (22.53 billion rows/s., 180.25 GB/s.)|
+| SELECT count(number) FROM numbers_mt(100000000000)           | 2.125 s.<br /> (47.07 billion rows/s., 376.53 GB/s.)|
+| SELECT sum(number+number+number) FROM numbers_mt(100000000000) | 17.169 s.<br /> (5.82 billion rows/s., 46.60 GB/s.)|
+| SELECT sum(number) / count(number) FROM numbers_mt(100000000000) | 3.696 s.<br /> (27.06 billion rows/s., 216.45 GB/s.)|
+| SELECT sum(number) / count(number), max(number), min(number) FROM numbers_mt(100000000000) | 8.348 s.<br /> (11.98 billion rows/s., 95.83 GB/s.)|
+| SELECT number FROM numbers_mt(10000000000) ORDER BY number DESC LIMIT 10 | 3.164 s.<br /> (3.16 billion rows/s., 25.28 GB/s.)|
+| SELECT max(number), sum(number) FROM numbers_mt(1000000000) GROUP BY number % 3, number % 4, number % 5 LIMIT 10| 1.657 s.<br /> (603.62 million rows/s., 4.83 GB/s.)|
 
 !!! note "Notes"
 
-    DatafuseQuery system.numbers_mt is <b>16-way</b> parallelism processing, [gist](https://gist.github.com/BohuTANG/8c37f5390e129cfc9d648ff930d9ef03)
+    DatafuseQuery system.numbers_mt is <b>16-way</b> parallelism processing, [gist](https://gist.github.com/BohuTANG/ab211a47c1493b9ecd01bf99a7731037)
 
 <figure>
   <img src="https://datafuse-1253727613.cos.ap-hongkong.myqcloud.com/datafuse-avg-100b.gif"/>