Qualcomm Announced the Strongest 10nm 48 core 120W ARM Centriq 2400 Processor


Qualcomm today, has announced the world’s first and only commercial Centriq 2400 Series 10-nm server.

The Qualcomm Centriq 2400 processor family is based on a Samsung 10-nanometer FinFET process designed as a SoC platform-level solution with 18 billion transistors integrated into 398mm2.

Coming on specifications, the series included three processors, where Centriq 2460 is presented as the flagship, it contains 48 high-performance 64-bit single-threaded Falkor self-developed core (based on ARMv8, Qualcomm fifth generation of autonomous architecture), the basic frequency of 2.2GHz, Up to 2.6 GHz

Each of the two cores shares 512 KB of L2 Cache, up to 60MB and up to 120W of thermal design power (40 cores of Centriq 2434 at 110W).

In internal design, each core is connected by a two-way, segmented, closed-loop bus with a total bandwidth of 250 Gbps to avoid performance bottlenecks at full load.

The processor support DDR4-2667 memory, the maximum capacity of 768GB, 6-channel, 32 PCIe Gen3 channels and 6 PCIe controllers.

A number of cloud service providers and technology companies attended the launch event and demonstrated the Arm data centre applications supported by the Centriq 2400 including – Alibaba, American Megatrends Inc., Arm, Cadence Design Systems, Canonical, Chelsio Communications, Cloudflare, Excelero, HCI, Illumina, LinkedIn, MariaDB, Mellanox, Microsoft Azure , MongoDB, Netronome, Packet, Red Hat, ScyllaDB, 6WIND, Samsung, Solarflare, CoreRay, SUSE, Uber and Xilinx.

Some of the key applications and use cases demonstrated today include:

  • Web front end to end support, HipHop Virtual Machine NoSQL databases, including MongoDB, Varnish, and Scylladb.
  • Cloud platform management and automation including Kubernetes, Docker and metal as a service Data Analysis, including Apache Spark.
  • In-depth learning reasoning, Network function virtualization, Video and image processing speed up, Multi-core electronic design automation.
  • High throughput computational bioinformatics.
  • Neural hierarchy network, OpenStack platform.
  • Scale-out Server SAN with NVMe support.
  • Server-based network load shedding.

For the latest tech news and updates follow TechnoCodex on Facebook, TwitterGoogle+. Also, if you like our efforts, consider sharing this news with your friends, this will encourage us to bring more interesting stories for you.

Get real time updates directly on you device, subscribe now.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. AcceptRead More