中文 English

Week In Review: Design, Low Power

Synopsys buys performance optimization startup; measuring quantum performance; mobile ray tracing; financial results.

popularity

Business

Synopsys acquired Concertio, a provider of AI-powered performance optimization software. The acquisition will bolster Synopsys’ silicon lifecycle management platform SiliconMAX SLM with the addition of Concertio’s autonomous software agent that, when installed on the target system, continuously monitors the interactions between operating applications and the underlying system environment, learning about the behavior of the applications through reinforcement learning techniques. The optimization engine can then adapt and reconfigure the system dynamically and improve integration of AI at the edge with big data analytics in the cloud. Based in New York, New York, Concertio was founded in 2016.

Ansys released Q3 2021 financial results with revenue of $441.2 million, up 20% from the same quarter last year. “Our year-to-date results indicate we are tracking to our business model of double-digit growth with industry-leading margins. Looking towards the end of the year, we continue to see a robust deal pipeline and momentum in the business, bolstering our confidence to raise full-year financial guidance above and beyond the impact of our strong Q3 top-line performance,” said Ansys CFO Nicole Anasenes.

Rambus reported third quarter 2021 financial results with revenue of $81.3 million, an increase of 23.2%. “We are well positioned for continued profitable growth as demonstrated by this quarter’s record product revenue from memory interface chips. Strategically, we continue to scale the business as the integrations of AnalogX and PLDA are well underway with the new teams already contributing new products and design wins,” said Rambus CEO Luc Seraphin.

Seagate inked a multi-year agreement to use Ansys’ simulation solutions. Seagate has used Ansys’ tools including OptiSLang in the development of its Heat-Assisted Magnetic Recording (HAMR) technology.

XtremeEDA signed on as a Codasip Certified Design Services Company. XtremeEDA will provide services, consulting, and support to companies developing custom RISC-V processor designs using Codasip tools and cores.

Oculii deployed Ansys electromagnetic simulation solutions in development of its AI software and hardware for automotive radar systems. It used Ansys HFSS to optimize key elements of radar transmission, including antenna and sensor placement and performance, with a predictive accuracy of 80% to 90%.

IP

Imagination Technologies uncorked new GPU IP, IMG CXT. It includes the PowerVR Photon ray tracing architecture, which aims to bring ray tracing to applications such as smartphones, tablets, and automotive HMI platforms. IMG CXT also has 50% more compute, texturing, and geometry performance than the previous-generation GPU IP.

Truechip introduced network-on-chip (NoC) IP for RISC-V based chips supporting the TileLink protocol. It currently supports TileLink UL and UH conformance levels, with support for TL-C (cache coherency) planned. Primarily a verification IP provider, this is Truechip’s first silicon IP offering.

Quantum computing
IBM proposed a new metric to measure the performance of quantum systems. Called Circuit Layer Operations Per Second, or CLOPS, it indicates how many circuits can run on hardware in a given time and includes not just the speed of the quantum system but also its interaction with a classical system. IBM researchers wrote that the benchmark “requires execution of many instances of the model circuit with different parameters generated at runtime. Various parts of this hardware-software stack contribute to CLOPS, including the repetition rate of the quantum processor, the speed at which gates run, the runtime compilation, the amount of time it takes to generate the classical control instructions, and finally, the data transfer rate among all units.”

The U.S. and U.K. governments signed a joint statement pledging to work together on quantum information science and technology. The statement notes support for cooperative R&D and development of market opportunities. “By working together, we can broaden training opportunities, develop new applications for quantum technology, and think globally about how to maximize the benefits of these technologies for everyone,” said Eric Lander, Director of the U.S. Office of Science and Technology Policy. The U.S. National Institute of Standards and Technology (NIST) and the U.K. National Physical Laboratory (NPL) have already been collaborating on metrology research and standards for quantum technologies including next-generation atomic clocks and quantum sensors.

Q-CTRL said it has identified error-suppression techniques that increase the likelihood of quantum computing algorithm success over 2,500% on real hardware (IBM’s, in this instance). Its software modifies the quantum logic gates used to construct algorithms, without changing the test algorithm or hardware itself.



Leave a Reply


(Note: This name will be displayed publicly)