2020 Q2 Hyperledger Besu
Created by Grace Hartley, last modified by Gari Singh on Jun 11, 2020
Besu remains a strong project with a growing community. This quarter Besu has prioritized fostering the community around Besu and delivering its 1.5. Release.
Questions/Issues for the TSC
The Besu team doesn’t have any outstanding questions or issues. The team has been pleased with the community’s response to gaining its active status designation and the RocketChat updates that have made it more user friendly to join the chat channels.
Hyperledger Besu has completed eight bi-weekly patch releases (1.4.2 on 25 Mar, 1.4.3 on 8 Apr, 1.4.4 on 22 Apr, 1.4.5 RC1 on May 6, 1.4.5 RC2 on 12 May, 1.4.5 on 21 May, 1.4.6 RC1 on 29 May, 1.4.6 RC2 on 3 Jun).
Some functional improvements include:
- Introduced SecurityModule plugin API
- Changed the
user:groupon the Docker container to
- PlugIn APIs
- Onchain privacy groups with add and remove members.
Go to the Changelog for more details.
Overall Activity in the Past Quarter
A few high-level activities include:
- The Hyperledger Besu security audit was completed. Read the blog with a link to the findings here .
- Continue to run our bi-weekly contributor calls and grow the attendance.
- We have been using the Hyperledger tools consistently and with continued success. With the latest release of RocketChat, our team is very pleased because community members can see the chat channels without a LFID.
- Team completed transition to GitHub Issues from Jira issues
As a part of participating in the Hyperledger community, the Besu team has participated in the community by:
- Continuing to work on Besu’s support of Caliper, targeting to publish blog in July on our work with Caliper.
- Presented Hyperledger meet-ups, including Hyperledger Barcelona meet-up and Hyperledger Argentina meet-up in April
- Mark Wagner is leading the Hyperledger mentorship project to have Besu run on OpenShift
These plans remain fairly similar to our prior report published in late-March.
The project team remains currently working towards its 1.5 Release, scheduled for July 15, 2020. The 1.5 Release is expected to include the following features:
- Performance improvements: Performance is a high priority item for the Besu team at this time to ensure it’s a leading Ethereum space. Some of the improvements will include block propagation, block product and validation, transaction pool management, and JSON RPC query response time.
- Privacy Improvements
Preliminary Mainnet Berlin support.
The Besu team is also continuing to focus on finalizing and publishing performance metrics with its work on Hyperledger Caliper. The publishing of the metrics was slightly delayed since last quarter because there remains additional work for the team to get the range of metrics and tests it requires. The team is currently working on this task. We still believe publishing these performance metrics will help provide the community additional context on the project’s usability for a variety of use cases.
- Similar to last quarter, Besu is also continuing to engage with its community and grow the diversity of its maintainer and contributor base.
Our maintainer diversity remained fairly consistent from the prior quarter. We continue to have maintainers from four different organizations.
The four organizations include:
- ConsenSys (PegaSys)
- Splunk (maintainer was formerly at Machine Consultancy)
The new maintainers this quarter include:
- Karim Taan (PegaSys)
We had a couple of maintainers leave ConsenSys in the last quarter.
The maintainers breakdown is currently:
- 21% non-PegaSys - This is an increase from 13% in the prior quarter.
Commits from 2020-03-21 to 2020-06-04 : 318
Committers from 2020-03-21 to 2020-06-04 : 39 (10 non-PegaSys)
Identified Orgs 2020-03-21 to 2020-06-04 : 7 + 4 independent
- ✅ Angelo De Caro
- ✅ Arnaud J Le Hors
- ✅ Christopher Ferris
- ✅ Dan Middleton
- ✅ Gari Singh
- ✅ Hart Montgomery
- ✅ Mark Wagner
- ✅ Nathan George
- ✅ Swetha Repakula
- ✅ Tracy Kuhrt
- ✅ Troy Ronda
Looks as if you are making great progress. Thanks
Posted by ChristopherFerris at Jun 10, 2020 19:19
" our team is very pleased because community members can see the chat channels without a LFID."
Posted by firstname.lastname@example.org at Jun 10, 2020 21:07
There was a bugfix in the new version of chat to allow un-authenticated users to read all public channels.
Posted by ryjones at Jun 10, 2020 21:11
I agree with Chris here, Great Report!.
Would love to hear more about the performance work on a Performance and Scalability Working Group call when your team is ready, but I may be biased. Until then, let the PSWG know if we can do anything to help. BTW, Vipin is always great for quotes on blog posts, etc.
Posted by mwagner at Jun 11, 2020 01:32
The subtlety of the bug fix is we can now link directly to our ch"at room whereas before we had to link to the #general room and tell new users to go to the #besu room. That combined with the support of making lightweight LFIDs from github, gmail, and facebook accounts it eliminates the friction for casual users to (a) see whats being discussed and (b) join the discussion.
Posted by shemnon at Jun 11, 2020 04:45
For this quarter the performance improvements were all below the architecture level. Mainly doing some profiler driven optimizations and using native encryption libraries, such as the hand-coded assembly of bitcoin-core secp256k1 cure.
In the next quarter or two there may be some interesting architectural changes to talk about that are preparatory for stateless ethereum (fka Eth 1.x) and Eth2 phase 2, both of which I expect to have implementation impacts into the Besu codebase.
Posted by shemnon at Jun 11, 2020 04:47