As we approach the physical limits of semiconductors, new technologies are required to develop advanced chips. New materials, device types and more efficient architectures and packaging are necessary to satisfy the requirements of advanced end-consumer devices. This is also resulting in higher complexity in chip manufacturing, making semiconductor test an integral part of high-volume production processes.
“Semiconductor manufacturing is only getting more complicated,” said Regan Mills, vice president of marketing and general manager, Teradyne’s Semiconductor Test Division. “The industry has been able to innovate and enable Moore’s Law for years, and chips are getting more powerful at lower costs. To continue that path requires a collection of simultaneous innovations in process and packaging that increase complexity rapidly and requires a shift in our thinking, as a community, toward analytics.”
When performing test, customers need to receive the right data from the tester in a format they can understand and use, and they want the ability to make improvements and optimizations in real-time. In addition, all data must be secure and accessible to only the right people.
To meet these requirements, Teradyne recently launched its open-architecture Archimedes analytics solution that brings real-time analytics to semiconductor test. Easily integrating with widely used data analytics platforms, including proteanTecs and PDF Solutions, Teradyne is extending the power of machine learning and artificial intelligence from established data analytics companies to a secure, real-time environment.
The open-architecture Archimedes analytics solution brings real-time analytics to semiconductor test. (Source: Teradyne)
Real-time analytics
Semiconductor testing on automated test equipment (ATE) produces large amounts of data during validation and high-volume manufacturing test processes. Current analytics solutions use diverse data formats, specifications and interface requirements, and they need on-tester agents for compatibility with automated test solutions, resulting in increased support and application-engineering costs, said Teradyne.
This approach, due to its overhead and reliance on the cloud, also may slow down tester operations and expose critical test data.
The Archimedes analytics solution eliminates the need for on-tester agents and delivers real-time analytics to semiconductor test, optimizing test flow and yield while reducing both costs and security risks.
The solution incorporates data analytics, artificial intelligence and machine learning into the test solution. According to Teradyne it delivers a secure stream of real-time data, as well as near-zero test time, improved yield and quality and rapid time to analysis. It identifies failures and their root causes, allowing for corrective actions to be taken before and during high-volume manufacturing. Archimedes’ bi-directional data stream enables real-time analytics and then pushes learnings back to the tester for immediate test program optimizations.
An integral part of the Archimedes analytics solution is the UltraEdge2000 high-performance parallel-compute platform with millisecond latency, which ensures that heavy computational processes take place at the edge of the network without slowing down the tester and test flow.
“When it comes to security, there are two important things. The first is: who owns the data inside the fab and who has access to it,” Mills said. “The truth is that everyone wants to protect their IP. The second thing is that the data needs the right structure and needs to remain unfiltered to be used by the downstream systems.”
Archimedes addresses three aspects of security:
- Can you trust the hardware and software being deployed in your environment?
- Can you trust the data that is delivered?
- Can the data be secured so that only the right people have access?
Teradyne said that Archimedes is a natively secure environment, eliminating risks associated with cloud-based solutions and the need for on-tester agents. The UltraEdge2000 zero-trust environment ensures only the right people have access to the data.
What it means
“When we talk about an open environment, we want to make sure that our products and services integrate cleanly with our customers’ enterprise solutions, which they either have now or will have at some point,” said Mills. “Working with industry associations like SEMI, which has spearheaded the SEMI Data-AI initiative, ensures the entire ecosystem is moving in the same direction, ultimately moving the industry forward.”
Teradyne said its UltraEdge2000 brings these capabilities to real-time implementation. In one case, for example, a customer reduced its product development timeline plus achieved a 14% reduction in power consumption with the combined solution.
“Using data analytics in an aggregated manner will keep the semiconductor industry moving in the right direction,” Mills said. “We’re not replacing engineers. We’re using analytics to help engineers make better decisions more quickly. I like to call this AI-accelerated insights,” he added.
Advertisement
