Week In Review: Design, Low Power

Arm acquired Treasure Data, a provider of enterprise data management solutions. Along with Mbed Cloud and the recent acquisition of Stream, which provides connectivity and device management, it will form the basis of Arm’s new IoT management platform. Treasure Data was founded in 2011. Terms of the deal were not disclosed, although Bloomberg reported the price at $600 million.

UltraSoC now supports SEGGER J-Link debug probes within its integrated SoC monitoring and analytics environment. The probes support processor platforms including RISC-V, and both current and legacy Arm cores.

The MIPI Alliance initiated development of a physical layer (up to 15m) specification targeted for ADS, ADAS and other surround sensor applications. Development of the MIPI A-PHY physical layer specification is underway to meet 12-24Gbps, and the organization is gathering requirements to support speeds over 48Gbps for display and other use cases. MIPI A-PHY v1.0 is expected to be available to developers in late 2019, with the first vehicles using A-PHY components expected in 2024. MIPI says the specification will optimize wiring, cost and weight requirements, as high-speed data, control data and optional power share the same physical wiring.

Rambus reported second quarter 2018 financial results with revenue of $56.5 million. The company saw a net loss per share of $0.14 on a GAAP basis and a non-GAAP net loss per share of $0.03. This year, new accounting standards were adopted; under the previous standard, revenue was $98.8 million, up 4.3% from the second quarter 2017 (and 9% higher excluding the impact of the Lighting Division which was wound down in the first quarter). GAAP net income per share was $0.13, up 550% from $0.02, and non-GAAP net income per share was $0.21, up 50% from $0.14 in Q2 2017.

Events
Flash Memory Summit: August 7-9 in Santa Clara, CA. The conference will feature discussion of NVMe, NVMe-oF, Persistent Memory, advanced memory technologies, and key Open Source software topics.

Hot Chips: August 19-21 in Cupertino, CA. The symposium for high-performance chips will feature a number of sessions on machine learning architectures as well as discussions of hardware security methods, blockchain, and acceleration.