翻譯出處： https://academy.datastax.com/planet-cassandra/nosql-performance-benchmarks

Apache Cassandra NoSQL效率標(biāo)準(zhǔn)

Apache Cassandra? is a leading NoSQL database platform for modern applications. By offering benefits of continuous availability, high scalability & performance, strong security, and operational simplicity — while lowering overall cost of ownership — Cassandra has become a proven choice for both technical and business stakeholders. When compared to other database platforms such as HBase, MongoDB, Redis, MySQL and many others, the linearly scalable database Apache Cassandra? delivers higher performance under heavy workloads.

Apache Cassandra?? 是領(lǐng)先的NoSQL應(yīng)用數(shù)據(jù)庫平臺. 提供可持續(xù)應(yīng)用能力简卧，高伸縮和性能，強大安全性和操作簡單—同時降低學(xué)習(xí)成本—Cassandra 已經(jīng)成為技術(shù)和業(yè)務(wù)兩者的最佳選擇。對比其他數(shù)據(jù)庫如Hbase，MongoDB,Redis谢澈，Mysql和更多其他數(shù)據(jù)庫，這個可伸縮數(shù)據(jù)庫 Apache Cassandra? 在高負(fù)載的情況下能提供高性能服務(wù)糊余。

The following benchmark tests provide a graphical, ‘a(chǎn)t a glance’ view of how these platforms compare under different scenarios. ?When selecting a database it is critically important to understand your use case and find the right fit. Below you will find the following three bechmarks; taking a look at write/read performance and performance at scale:?

下面的基準(zhǔn)測試提供了一個圖形對比霍殴，‘對比’ 可以看出在不同場景下面數(shù)據(jù)庫的比較。在選擇的數(shù)據(jù)庫時帝簇，了解你應(yīng)用數(shù)據(jù)庫使用場景和選擇適合的數(shù)據(jù)庫是重要的徘郭。你會發(fā)現(xiàn)以下三種情況：看下寫/讀的情況和掃描情況：

University of Toronto Benchmark

多倫多大學(xué)的標(biāo)準(zhǔn)

Netflix: Benchmarking Apache Cassandra Scalability

Netflix公司：Apache Cassandra 可擴展性能

End Point Benchmark Configuration and Results

最后得到標(biāo)準(zhǔn)的配置和結(jié)果

University of Toronto NoSQL Database Performance

多倫多大學(xué) NoSQL數(shù)據(jù)庫性能

Engineers at the University of Toronto, in 2012, conducted a thorough benchmarking analysis of various NoSQL platforms including: Apache Cassandra, HBase, MySQL, Redis and Voldemort. The testing was extremely thorough and included a view into performance under varying workloads.

在2012年，多倫多大學(xué)的工程師們丧肴，對于各種的NoSQL數(shù)據(jù)庫做了全面的性能測試：Apache Cassandra残揉，Hbase，MySQL芋浮，Redis 和 Voldemort.測試得非常全面抱环，包括在不同的極端工作負(fù)載條件下的性能統(tǒng)計。

For a look at the details behind this analysis as well as a complete write up of the benchmark configurations used, the white paperSolving Big Data Challenges for Enterprise Application Performance Managementprovides all of the insight from this test. Overall their results identified Apache Cassandra the “clear winner throughout our experiments”.

來查看一下分析細(xì)節(jié)背后的一個完整的基準(zhǔn)測試纸巷，《解決企業(yè)應(yīng)用程序性能管理的大數(shù)據(jù)挑戰(zhàn)》的白皮書提供了這個測試的所有見解.總體來說镇草，Apache Cassandra 是“整個實驗當(dāng)中明顯的贏家”.

A summary of throughput and latency results are available here.

吞吐量和延遲結(jié)果在匯總?cè)缦?

Throughput for workload Read/Write

吞吐量在工作中的讀/寫

Throughput for workload Read/Scan/Write

工作中的吞吐量 讀/掃描/寫

Read latency for workload Read/Write

工作中的讀等待 讀/寫

Write latency for workload Read/Write

工作中的寫操作 讀/寫

If this benchmarking data from University of Toronto is interesting,take a 10 minute Cassandra walkthroughand learn more.

多倫多大學(xué)對于標(biāo)準(zhǔn)數(shù)據(jù)得到的有趣結(jié)果，十分鐘 Cassandra 入門和學(xué)習(xí)?

Netflix

Netflix decided to run a test designed to validate their tooling and automation scalability as well as the performance characteristics of Cassandra. The results of their testing are provided below. For a more thorough write up of the Netflix testing process including configuration settings and commentary, visit their tech blog post titledBenchmarking Cassandra Scalability on AWS – Over a million writes per second.

Netflix 決定運行一個測試來驗證它們的工具和自動化可伸縮性以及Cassandra的性能特性.這測試結(jié)果提供如下:有關(guān)Netflix測試過程的更詳細(xì)的描述瘤旨，包括設(shè)置配置和評論梯啤，請訪問他們的技術(shù)博客，標(biāo)題為《對AWS的Casdand可伸縮性進(jìn)行基準(zhǔn)測試-每秒超過一百萬次寫入》.

End Point Benchmark Configuration and Results Summary

最終基本配置和結(jié)果摘要

End Point, a database and open source consulting company, benchmarked the top NoSQL databases — Apache Cassandra, Apache HBase, and MongoDB — using a variety of different workloads on Amazon Web Services EC2 instances. This is an industry-standard platform for hosting horizontally scalable services such as the NoSQL databases that were tested. In order to minimize the effect of AWS CPU and I/O variability, End Point performed each test 3 times on 3 different days. New EC2 instances were used for each test run to further reduce the impact of any “l(fā)ame instance” or “noisy neighbor” effect on any one test.

最終存哲，一個數(shù)據(jù)庫和開源咨詢公司条辟，檢測到最好的NoSQL數(shù)據(jù)庫—??Apache Cassandra, Apache HBase, and MongoDB — 運用了大量的不同工作負(fù)載在亞馬遜Web服務(wù) EC2實例上黔夭。這是一個可伸縮服務(wù)的行業(yè)標(biāo)準(zhǔn)平臺，如被測試的NoSQL數(shù)據(jù)庫.為了盡量減少AWS CPU和I/O可變性的影響羽嫡。結(jié)束點在3個不同的時間進(jìn)行了3次測試.新EC2實例能運用每次測試執(zhí)行進(jìn)一步減少任何“差勁的實例”或“嘈雜的鄰居” 效率對任何一個測試的影響本姥。

A summary of the workload analysis is available below. For a review of the entire testing process with testing environment configuration details, thebenchmarking NoSQL databases white paperby End Point is available.

下面是工作負(fù)載分析的總結(jié).對于具有測試環(huán)境配置細(xì)節(jié)的整個測試過程的回顧,可以使用對《NOSQL數(shù)據(jù)庫白皮書》的最終基準(zhǔn)測試。

Goals for the Tests

成功的測試

Select workloads that are typical of today’s modern applications

選擇當(dāng)今現(xiàn)代應(yīng)用中典型的工作負(fù)載杭棵。

Use data volumes that are representative of ‘big data’ datasets that exceed the RAM capacity for each node

運用數(shù)據(jù)量可以代表“大數(shù)據(jù)” 數(shù)據(jù)集并超過每個節(jié)點的RAM容量婚惫。

Ensure that all data written was done in a manner that allowed no data loss (i.e. durable writes), which is what most production environments require

確保所有數(shù)據(jù)已經(jīng)寫入以允許數(shù)據(jù)丟失的方式(持久化).這也是大多生產(chǎn)環(huán)境的需要。

Tested Workloads

測試負(fù)載

The following workloads were included in the benchmark:

以下工作負(fù)載中包括在基準(zhǔn)中：

Read-mostly workload, based on YCSB’s provided workload B: 95% read to 5% update ratio

更多讀取工作量魂爪，基于YCSB’s 提供工作負(fù)載 B：95% 讀 5% 更新率

Read/write combination, based on YCSB’s workload A: 50% read to 50% update ratio

基于YCSB’s的工作負(fù)荷A：50% 讀取50% 更新率

Read-modify-write, based on YCSB workload F: 50% read to 50% read-modify-write

讀-修改-寫先舷，基于基于YCSB’s的工作負(fù)荷 F：50% 讀 50% 讀-修改-寫

Mixed operational and analytical: 60% read, 25% update, 10% insert, and 5% scan

最大操作和解析： 60% 讀，25% 修改滓侍，10%插入和5%掃描

Insert-mostly combined with read: 90% insert to 10% read ratio

插入和讀：90% 插入到10%讀

Throughput Results

吞吐量結(jié)果：

Get started with the best distribution of Apache Cassandra?

從Apache Cassandra 的最佳發(fā)行開始