Method and device suitable for thousand-gear market based on programmable device

Abstract

The invention discloses a method and device suitable for thousand-gear market based on a programmable device, and the method comprises the steps: designing an FPGA implementation architecture in an FPGA acceleration board card F37X, enabling the FPGA implementation architecture to comprise a read-write control unit, a cache structure and a DDR / HBM, enabling the read-write control unit to carry outthe read-write one-by-one delegation, and enabling the DDR / HBM to store all delegation information; in the cache structure, two paths of parallel read-write caches with configurable depths are adopted, the two paths of parallel read-write caches respectively cope with random one-by-one delegated reading inside and outside a read-write window, and when one-by-one transaction information comes, thetwo paths of parallel read-write caches inquire delegated prices according to message record numbers. Through the above mode, the message record number query hit rate can be improved, the entrustingprice reading delay is reduced, and the overall system processing performance is improved.

Description

technical field [0001] The invention relates to the field of financial markets, in particular to a method and device based on a programmable device suitable for thousands of market quotations. Background technique [0002] Thousands of quotes in Shenzhen Stock Exchange has become one of the core functions of Level-2 market quotes, which is traditionally implemented by software. However, with the vigorous development of my country's financial market, traditional software technology or software-centric hardware and software acceleration technology is difficult to meet The current real-time response requirements, FPGA-based hardware and software co-acceleration technology is an effective way to solve this problem. [0003] The Shenzhen Stock Exchange Level-2 market includes two types of data, order by order and transaction by transaction, so the receiver can use the data and trading rules to reconstruct the market by itself. At present, the mainstream solution is still software...

AI Technical Summary

Problems solved by technology

At present, the mainstream solution is still software reconstruction. The main problem is to use data structures such as linked lists, balanced trees, and hash tables to store market information. The processing capacity is far from meeting the real-time data requirements of the current market. In addition, due to the randomness of transaction by transaction, As a result, the cache hit rate dropped sharply when entrusting each order to find, and hash search and other methods had to be used. However, the problems brought about were hash conflicts and huge delays in memory reading, which could not meet the extremely low-latency requirements in the field of financial acceleration.

Images