|
September Presto TSC Meeting Postponed
Gale McCommons <gmccommons@...>
|
|
|
|
[Proposal] Adding a new top-level project in Presto on GitHub
rongrong100@...
As part of the modernization effort in Facebook, we want to develop a new parser front end. The goals for this new project includes: * SQL16 compliance * New language extensions for rich types, streaming, graph process, etc. * Fully resolved AST * AST rewriter framework and optimizations * Fully compatible with current presto-parser
We think a stand-alone project is better because we want to push this to be a more general SQL standard for other SQL-based query processing frameworks as well. But this will still be under the presto org as a Presto Foundation project and we want to ensure this is fully compatible with existing Presto grammar. Please reply if you have any questions or concerns on this project.
Thanks! Rongrong
|
|
|
|
[Proposal] Adding presto-query-predictor as a new top-level project in Presto on GitHub
zluo@...
Hi, We'd like to add presto-query-predictor as a new top level project in Presto. The Presto query predictor introduces machine learning techniques to provide a quick estimate of resource usage (CPU time and peak memory bytes) of a Presto query. It is achieved by training ML models from historical Presto logs. At Twitter, the project helped with load balancing, traffic management, etc. Currently, we have open-sourced the project in a separate branch in the twitter-fork presto repo. https://github.com/twitter-forks/presto/tree/query-predictor/presto-query-predictor The documentation is served at https://chunxutang.github.io/presto-query-predictor-docs/ The codebase is written in Python. Why create a new repo for the project? Since open source, we have received interests/questions/feature-requests from multiple Presto developers/users. Keeping the project in twitter forked presto branch brings up troubles in: code sharing, Python build process, and feature support. For example, we don’t have a specific GitHub issue tracker for the project, which makes it not convenient for us to answer questions or feature requests. It’s also cumbersome to create a unified build process for the Python module. By creating a new repo under Presto umbrella, we could get: A unified platform to answer questions and feature requests. A primary repo/branch for releases and Python package maintenance. An easily discovered codebase for viewing and sharing. More collaboration with the open-source community of introducing ML techniques to the Presto ecosystem. Please reply if you have any questions or concerns on this project. Thanks, Zhenxiao
|
|
|
|
Re: [Proposal] Adding a new top-level project in Presto on GitHub
Maria Basmanova
Rongrong,
do you plan to use Java or C++ for this new project?
-Masha
From: <presto-tsc@...> on behalf of "rongrong100@..." <rongrong100@...>
As part of the modernization effort in Facebook, we want to develop a new parser front end. The goals for this new project includes: * SQL16 compliance * New language extensions for rich types, streaming, graph process, etc. * Fully resolved AST * AST rewriter framework and optimizations * Fully compatible with current presto-parser
We think a stand-alone project is better because we want to push this to be a more general SQL standard for other SQL-based query processing frameworks as well. But this will still be under the presto org as a Presto Foundation project and we want to ensure this is fully compatible with existing Presto grammar. Please reply if you have any questions or concerns on this project.
Thanks! Rongrong
|
|
|
|
Re: [Proposal] Adding a new top-level project in Presto on GitHub
rongrong100@...
Our current plan is to use Java. We decided to go with Java rather than C++ because it’s easier to integrate with presto coordinator and java APIs for metadata are well supported while c++ APIs are lacking. Thanks! Rongrong
On Thu, Sep 24, 2020 at 4:16 AM Masha Basmanova <mbasmanova@...> wrote:
--
Rongrong
|
|
|
|
Nominating Venki Korukanti as a Committer
Nezih Yigitbasi
Hey everyone,
Venki (GitHub handle: vkorukanti) has been contributing to the PrestoDB community as a probationary committer since Feb. Here is a highlight of his contributions: - Commits: 31 commits 8,778 ++ 802 -- - Major features - vectorized parquet reader - nested column project and filter pushdown for Parquet (in collaboration with Zhenxiao @ Twitter) - Bug fixes in various areas: planning, execution, and subfield pushdown - Community - code reviews and helping users with Parquet/Hive related issues - Presto meetup talk (Oct 17, 2019) - Encouraging and providing guidance to Uber Presto/Data team members to contribute to PrestoDB and helps the community grow Per the TSC charter "A Contributor may become a Committer by a majority approval of the existing Committers.", therefore, each TSC member has one vote to cast. Please reply with +1/-1 if you approve/disapprove onboarding Venki as a committer. Thanks! Nezih
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
zluo@...
+1
Hey everyone,
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
jamessun@...
Personally, I would love to observe at least another larger PR to show Venki is ready for committership. The major reason behind is due to the design on https://github.com/prestodb/presto/pull/14834 and https://github.com/prestodb/presto/pull/15140. These two PRs to me feel a bit more design flaws.
Thanks
James
From: <presto-tsc@...> on behalf of "zluo via lists.prestodb.io" <zluo=twitter.com@...>
+1
On Fri, Sep 25, 2020 at 10:26 AM nezihy via
lists.prestodb.io <nezihy=fb.com@...> wrote:
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
jamessun@...
Just realized I got the wrong person. Re-checked the Venki’s PRs. I’m supportive.
From: <presto-tsc@...> on behalf of "jamessun via lists.prestodb.io" <jamessun=fb.com@...>
Personally, I would love to observe at least another larger PR to show Venki is ready for committership. The major reason behind is due to the design on https://github.com/prestodb/presto/pull/14834 and https://github.com/prestodb/presto/pull/15140. These two PRs to me feel a bit more design flaws.
Thanks
James
From: <presto-tsc@...> on behalf of "zluo via lists.prestodb.io" <zluo=twitter.com@...>
+1
On Fri, Sep 25, 2020 at 10:26 AM nezihy via
lists.prestodb.io <nezihy=fb.com@...> wrote:
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Wenlei Xie
+1
Hey everyone, --
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Tim Meehan
After scanning through Venki’s commits and issues I am supportive.
From: <presto-tsc@...> on behalf of "wenlei.xie@..." <wenlei.xie@...>
+1
On Fri, Sep 25, 2020 at 10:26 AM nezihy via
lists.prestodb.io <nezihy=fb.com@...> wrote:
-- Best Regards,
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Shixuan Fan
+1 after skimming through several Venki’s pull requests
From: <presto-tsc@...> on behalf of "Tim Meehan via lists.prestodb.io" <tdm=fb.com@...>
After scanning through Venki’s commits and issues I am supportive.
From: <presto-tsc@...> on behalf of "wenlei.xie@..." <wenlei.xie@...>
+1
On Fri, Sep 25, 2020 at 10:26 AM nezihy via
lists.prestodb.io <nezihy=fb.com@...> wrote:
-- Best Regards,
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Andrii Rosa
+1 after skimming through Venki’s pull requests
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
rongrong100@...
+1
On Wed, Sep 30, 2020 at 7:40 PM Andrii Rosa <andriyrosa@...> wrote:
--
Rongrong
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Nezih Yigitbasi
Thank you all. We have a majority approval here and we can conclude the vote. Congrats, Venki!
Cheers,
Nezih
|
|
|
|
Re: Nominating Venki Korukanti as a Committer
Girish Baliga
Thanks everyone! Congrats, Venki!
|
|
|
|
Nominating Bin Fan as a Committer
jamessun@...
Dear TSC members,
Bin Fan from Alluxio has been contributing to Presto community in the past 10 months. Here, I nominate Bin as a committer with supporting docs listed below. Per the TSC charter "A Contributor may become a Committer by a majority approval of the existing Committers.", therefore, each TSC member has one vote to cast. Please reply with +1/-1 if you approve/disapprove onboarding Bin as a committer.
===== Bin's contributions to the Presto Community ===== Merged Pull Requests to Presto Source Code:
Major contribution:
Other pull requests:
Presto related tutorial or blogs:
Presto related public presentations:
Community service:
|
|
|
|
Re: Nominating Bin Fan as a Committer
Wenlei Xie
+1 after skimming through several PR and reviewing the design/implementation challenges. I would also like to thank Bin (together with Rohit and James) for making RaptorX happen. This brings the latency of shared-nothing architecture (Raptor) to the disaggregated storage architecture (RaptorX with Hive Connector) via carefully engineered cache (Alluxio).
--
|
|
|
|
Re: Nominating Bin Fan as a Committer
Shixuan Fan
+1. I’ve been working with Bin actively and would vouch for Bin. Alluxio integration makes RaptorX effort much easier, and there are more features like cache quota to come :)
From: <presto-tsc@...> on behalf of "jamessun via lists.prestodb.io" <jamessun=fb.com@...>
Dear TSC members,
Bin Fan from Alluxio has been contributing to Presto community in the past 10 months. Here, I nominate Bin as a committer with supporting docs listed below. Per the TSC charter "A Contributor may become a Committer by a majority approval of the existing Committers.", therefore, each TSC member has one vote to cast. Please reply with +1/-1 if you approve/disapprove onboarding Bin as a committer.
===== Bin's contributions to the Presto Community ===== Merged Pull Requests to Presto Source Code:
Major contribution:
Other pull requests:
Presto related tutorial or blogs:
Presto related public presentations:
Community service:
|
|
|
|
Re: Nominating Bin Fan as a Committer
rongrong100@...
+1 Rongrong
|
|
|