Procella: Unifying serving and analytical data at YouTube

被引：16

作者：

Chattopadhyay, Biswapesh ^{[1
]}

Dutta, Priyam ^{[1
]}

Liu, Weiran ^{[1
]}

Tinn, Ott ^{[1
]}

Mccormick, Andrew ^{[1
]}

Mokashi, Aniket ^{[1
]}

Harvey, Paul ^{[1
]}

Gonzalez, Hector ^{[1
]}

Lomax, David ^{[1
]}

Mittal, Sagar ^{[1
]}

Ebenstein, Roee ^{[1
]}

Mikhaylin, Nikita ^{[1
]}

Lee, Hung-ching ^{[1
]}

Zhao, Xiaoyan ^{[1
]}

Xu, Tony ^{[1
]}

Perez, Luis ^{[1
]}

Shahmohammadi, Farhad ^{[1
]}

Bui, Tran ^{[1
]}

Mckay, Neil ^{[1
]}

Aya, Selcuk ^{[1
]}

Lychagina, Vera ^{[1
]}

Elliott, Brett ^{[1
]}

机构：

[1] Google LLC, Mountain View, CA 94043 USA

来源：

PROCEEDINGS OF THE VLDB ENDOWMENT | 2019年 / 12卷 / 12期

关键词：

TIME;

D O I：

10.14778/3352063.3352121

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Large organizations like YouTube are dealing with exploding data volume and increasing demand for data driven applications. Broadly, these can be categorized as: reporting and dashboarding, embedded statistics in pages, time-series monitoring, and ad-hoc analysis. Typically, organizations build specialized infrastructure for each of these use cases. This, however, creates silos of data and processing, and results in a complex, expensive, and harder to maintain infrastructure. At YouTube, we solved this problem by building a new SQL query engine - Procella. Procella implements a super-set of capabilities required to address all of the four use cases above, with high scale and performance, in a single product. Today, Procella serves hundreds of billions of queries per day across all four workloads at YouTube and several other Google product areas.

引用

页码：2022 / 2034

页数：13

共 40 条

[1]

Abadi D.J., 2006, P 2006 ACM SIGMOD IN, P671, DOI [DOI 10.1145/1142473.1142548, 10.1145/1142473.1142548]

[2]

Abadi Daniel, 2013, Databases, V5, P197

[3]

Agrawal Sanjay, 2000, VLDB, P496

[4]

Ahmadi H., 2016, In-memory query execution in google bigquery

[5]

[Anonymous], ADD SPARK BIG DATA 1

[6]

[Anonymous], LAMBDA ARCHITECTURE

[7]

[Anonymous], EUROSYS

[8]

[Anonymous], CONFLUENCE COLUMN ST

[9]

[Anonymous], REAL TIME ANAL MASSI

[10]

[Anonymous], INFLUXDB TIM SER DAT

← 1 2 3 4 →