DeepBrain Chain Progress Report #105 02.16–02.28(2022)
Dear DeepBrain Chain community members,
Please see below for the latest update on the project.
I. Product Development Progress
GPU Cloud Platform:
- Fixed more than 20 bugs in the new version of the internal test for GPU cloud platform;
- Added the function of querying the list of virtual machines by data center and region;
- Added the function of Paypal payment for purchasing DBC;
- Optimization adjustment of deployment documents related to the cloud platform;
Distributed Computing Network:
- Realized DBC’s monitoring and API of various hardware parameter information, the latest version on github already supports these functions;
- Collation of documentation for monitoring;
- Modified the destination IP address of iptables, it now takes the IP address of the NIC device of the default route;
- Modified the mirroring center deployment method to adopt rsync-daemon mode, which is more secure;
- Fixed the problem that the DBC main program’s abnormal exit causes the download or upload process to not be terminated;
- Updated installation & upgrade scripts and related interface documents;
DeepBrain Chain Mainnet:
- On-chain updates:
1) Reduced the frequency of on-chain price acquisition;
2) Added testing of the reporting function;
3) Refactored the reporting code;
4) Now records machine ID and machine stash after quick reporting;
5) Added the script needed for reporting and added the description document;
6) Added some benchmark;
- On-Chain function fixes:
1) When the report is invalid and triggering re-reporting, increase the report ID instead of using the original ID for recording;
2) Fixed the problem of not refunding the pledge of the reporter after the report result expires in some cases;
3) Fixed the problem of not actively refunding the committee pledge when the pledge is cancelled;
II. Marketing Progress
- Progress of Galaxy Race Public Test:
DeepBrain Chain computing mainnet has been officially launched, as of February 28th, 2022 Beijing time, the latest data on-chain: a total of 16 computing pools participated, providing 1,329 GPUs, the total computing power value reached 278,368.65, these GPUs pledged a total of 133,315,000 DBC, the GPU rental rate reached 84.57%, and the accumulated rental income earned reached 183,390,698 DBC, adding DBC Korea Hyunjin-ICT pool from South Korea and Fastwalker — I — Deep Learning pool from the United States, see website for details.
- Preparations for the new council election have begun. To help candidates better present themselves to the community and to assist them in making campaign presentations to the community, the Foundation has improved the council campaign process, details of which can be found at: https://t.co/SnYaKdy1gN
- DBC ecosystem GPU cloud platform expansion has made positive progress: three new GPU cloud platforms have been added, namely hillstonehub (Korea), NexuiCom (USA), and Tianyuglobal.cloud (China), and FedML (USA) GPU cloud platform will be launched soon. The DBC Foundation’s support policy for GPU cloud platform has been released, which will help the development of GPU cloud platform in all aspects;
- The DBC Global Council has passed the proposal to restart the token switch, and will release specific information on the exchange process in the near future:https://www.reddit.com/r/DBC_Council/comments/rkrcpu/proposal_15_motion_17_open_the_swap_for_people/?utm_source=share&utm_medium=web2x&context=3
- GPU Cloud Platform Infrastructure Support Solution Released in Korean ‘’GPU클라우드 플랫폼에 대한 재단 지원 솔루션(2022.02) [출처] GPU클라우드 플랫폼에 대한 재단 지원 솔루션(2022.02)|작성자 DBC KOREA’’
- According to the data on the blockchain, the number of pledged DBCs on the mainnet reached 741.983 million, with a pledge rate of 17.1%, as detailed on: https://dbc.subscan.io/
- The proposal “Request for additional funding to support and help DBC deploy the NexuiCom GPU Cloud Platform in North America” was voted on and approved by the Board of Directors, and funding has been awarded, see https://www.reddit.com/r/DBC_Council/comments/ s9rg2k/proposal_gpu_cloud_computing/?utm_source=share&utm_medium=web2x&context=3
- To support the development of the GPU cloud platform, the Foundation has developed a relevant support program, which can be found at: https://docs.google.com/document/d/1N_6mtpxdWtHNHUkF3XvGy1LZ2CZcbUOV9sRYCNe30fA/edit#heading=h.17 zmq3xf0se4
III. Ecosystem Building
- Continued growth of incoming GPUs in the mainnet of computing power;
- The fourth model upgrade of the computing power value calculation is expected to be completed in March, and this model upgrade will add the computing value parameters of enterprise level GPUs such as A100, A40, V100, T4, etc., after which it will support the entry of enterprise level GPUs into the network;
- The construction of overseas computing power continues to be promoted, the South Korean pool will increase the deployment of computing power in March, and the pool in the Middle East and other places is expected to be officially launched in the second quarter of this year;
- In order to ensure the rental rate of your DBC computing nodes, please follow the recommended configuration and GPU model of the mainnet when launching online.
Contact information for joining the network:
— — — — — — — — — — — — — — — — — — — — — — —
Try our product as an AI user or a GPU power provider:
Main social media platform:
Telegram (English) ： https://t.me/DeepBrainChainGlobal
Official e-mail address: firstname.lastname@example.org
About DeepBrain Chain
DeepBrain Chain is the world’s first AI computing network powered by blockchain technology.
Our Mission: To connect global computing power, build a distributed high-performance computing power network, and construct computing power infrastructure for the era of 5G +AI.
Our Vision: Become the world’s largest computing network.
DeepBrain Chain’s computing power network can help AI practitioners/enterprises/universities/research institutions, cloud gaming, rendering, blockchain and other users to save on computing power cost, improve computing efficiency and enhance product experience.
The DeepBrain Chain Team