COBOL, one of the first programming languages to be used, dates back to about 1959. Production systems employ Over 800 billion lines of code, up from an estimated 220 billion in 2017, according to a 2022 poll. However, COBOL has a reputation for being a difficult-to-learn and ineffective language. COBOL is a difficult and expensive proposition for large organizations.
IBM today launched Code Assistant for IBM Z, which converts COBOL code into Java using a code-generating AI model. It is an effort to offer a fresh approach to the issue of modernizing COBOL software. Code Assistant for IBM Z will enter preview during IBM’s TechXchange conference in Las Vegas early in September. It will offer wide availability scheduled for Q4 2023.
According to IBM Research chief scientist Ruchir Puri, Code Assistant for IBM Z intends to help enterprises rework their mainframe applications while maintaining performance and security. Code Assistant is a code-generating model that may run locally in an on-premises setup or as a managed service in the cloud. CodeNet can comprehend more than 80 programming languages in addition to COBOL and Java. Code Assistant supports the whole application modernization life cycle and code transformation. It also aids developers in understanding, transforming, and validating the translated code in a modern architecture.
“Like any AI system, there might be unique usage patterns of an enterprise’s COBOL application that Code Assistant for IBM Z may not have mastered yet,” Puri said. “It’s essential that the code is scanned with state-of-the-art vulnerability scanners to ensure code security.”
CodeNet, according to Puri, has a big context window (32,000 tokens) to “capture the broader context” for “more efficient COBOL to Java transformation.” CodeNet is trained with 1.5 trillion tokens and contains 20 billion parameters.
Despite the risks, IBM undoubtedly views products like Code Assistant as crucial to its future development. Currently, 84% of IBM’s mainframe users run COBOL. The corporation sees the COBOL mainframe as a link to the vast hybrid computing systems that it hosts and supports.