ink!/pallet/solidity performance benchmarking phase 2

• Team Name: Talentica Software
• Payment Address: 0x8bd54ec34A35f3A2f668A33d9578b5C3A6b730dE
• Level: 1

Project Overview :

Proposal for Milestone 2 in the RFP titled implementation-benchmarking.

Overview

There are multiple ways to implement the logic in substrate i.e using pallets or ink smart contracts, or even writing solidity code and compiling it to WASM with the help of a solang compiler. We have to benchmark the performance metrics of the logic implemented using each of the above methods. We have already benchmarked the storage performance(basic data types) of each of the implementations as part of Milestone 1 and 2. Now, we will benchmark CPU-intensive logic, events emission and cross-contract calls across all the four implementations. We hope this will help new developers in deciding the best approach to implement the logic.

Project Details

We will employ the approach already taken to deliver Milestones 1 and 2 to benchmark CPU intensive task on all the four implementation strategies and also cross-contract calls within an implementation, wherever possible. We will use following tools to do so:

• Substrate Runtime Benchmarking Framework

Ecosystem Fit

This will help new developers to decide the best tool to implement the logic.

Team 👥

Team members

• Debasish Ray Chawdhuri - Blockchain and Cryptography Researcher
• Pankaj Mendki -Blockchain expert
• Nikhil Desai - Blockchain Developer https://github.com/Nikhil-Desai-Talentica
• Amit Singh - Blockchain Developer https://github.com/iamit-singh

Contact

• Contact Name: Nikhil Desai
• Contact Email: Nikhil.Desai@talentica.com
• Website: https://www.talentica.com/

Legal Structure

• Registered Address: Talentica Software Inc., 6200, Stoneridge Mall Road, Pleasanton, California 94588, USA.
• Registered Legal Entity: Talentica Software Inc.

Team's experience

We have already worked on implementation-benchmarking and delivered Milestones 1 and 2.

Team Code Repos

• implementation-benchmarking

• provenance-usecase

• simpleblockchain

• Nikhil-Desai-Talentica

Team LinkedIn Profiles (if available)

• Debasish Ray Chawdhuri - https://www.linkedin.com/in/debasish-ray-chawdhuri-4b444316/
• Pankaj Mendki - https://www.linkedin.com/in/pankaj-mendki-8576363/
• Nikhil Desai - https://www.linkedin.com/in/nikhil-desai-1209a38b/
• Amit Singh - https://www.linkedin.com/in/amit-singh-3637a898/

Development Status :

We have come up with a high-level implementation plan and will start implementing it soon.

Development Roadmap 🔩

Number	Objective	Deliverable	Time Estimate
1.		Milestone 3, Deliverable 1	2 days
2.		Milestone 3, Deliverable 2	2 days
3.		Milestone 3, Deliverable 3	1 day
4.		Milestone 3, Deliverable 4	2 days
5.		Milestone 3, Deliverable 5	3 days
6.		Milestone 3, Deliverable 6	3 day
7.		Milestone 3, Deliverable 7	3 days
8.		Milestone 3, Deliverable 8	3 days
9.		Milestone 3, Deliverable 9	1 days
10.		Milestone 3, Deliverable 10	1 days
11.		Milestone 3, Deliverable 11	1 days
12.		Milestone 3, Deliverable 12	1 days
13.		Milestone 3, Deliverable 0b, 0e	1 day

Overview

• Total Estimated Duration: 5 Weeks
• Full-Time Equivalent (FTE): 1
• Total Costs: 6,000 DAI

Milestone 3 - Compute intensive function and cross-contract call benchmarking

• Estimated duration: 5 Weeks
• FTE: 1
• Costs: 6,000 DAI

❗ The default deliverables 0a-0d below are mandatory for all milestones, and deliverable 0e at least for the last one.

Number	Deliverable	Specification
0a.	License	Apache 2.0 / GPLv3 / MIT / Unlicense
0b.	Documentation	We will provide both inline documentation of the code and and an article explaining how to use the tool to benchmark custom compute intensive functions and cross-contract calls along with a live demo if possible.
0c.	Testing and Testing Guide	We are building atop of the benchmarking framework that's part of the substrate-core. Some tests may be included when it comes to the functions being tested.
0d.	Docker	Not Applicable.
0e.	Article	We will publish an article that explains how each implementation technique performs, and when to choose which.
1.	Pallet CPU-intensive extrinsic	Extend a sample pallet to include a CPU-intensive extrinsic.
2.	Ink! CPU-intensive function	Extend a sample Ink! contract to include a CPU-intensive function.
3.	Solidity-WASM and Solidity-Native CPU-intensive function	Extend a sample Solidity contract to include a CPU-intensive function
4.	CPU-intensive benchmarks	Run the benchmarks on all these implementations and compare.
5.	Pallet cross-contract call	Add another pallet and invoke it from the sample pallet.
6.	Ink! cross-contract call	Add another Ink! contract and invoke it from the sample Ink! contract.
7.	cross-contract benchmarks	run cross-contract calls across different implementations and collect benchmarks.
8.	Solidity-WASM and Solidity-Native cross-contract call	Add another Solidity contract and invoke it from the sample Solidity contract. (Note: There are unresolved issues questioning the feasibility of cross-contract calls in Solidity-WASM and Solidity-Native. Nevertheless, an attempt will be made to see if it's possible.
9.	Pallet events	benchmark events emitted from the sample pallet.
10.	Ink! events	benchmark events emitted from the sample Ink! contract.
11.	Solidity-WASM and Solidity-Native events	benchmark events emitted from the Solidity-WASM and Solidity-Native contracts.
12.	Benchmark events across implementations	benchmark and record events emission across implementations.

We are planning to submit another grant application towards Milestone 4 where we discuss our approach to maintaining this work, using some software tools to automate the process wherever possible.

Additional Information ➕

Gautam Dhameja told us about the Grants program and encouraged us to apply to the same. We have already applied for and delivered Milestones 1 and 2 against implementation-benchmarking