Ampere^® Altra^® offers up to 80 cores at up to 3.0 GHz speed with sustained turbo performance. Each core is single threaded by design with its own 64 KB L1 I-cache, 64 KB L1 D-cache, and a huge 1 MB L2 D-cache, delivering predictable performance 100% of the time by fully eliminating the noisy neighbor challenge within each core.

Coherent mesh-based interconnect topology provides efficient bandwidth with 32 distributed home nodes and directory-based snoop filters to enable seamless and connectivity between the cores.

Supporting eight, 2DPC, 72-bit DDR4-3200 channels, the Ampere^® Altra^® processor offers high bandwidth and memory capacity of up to 4 TB per socket.

With leading power/core, and multi-socket support, Ampere^® Altra^® provides the scalability to maximize the number of servers per rack, unparalleled in the industry.

With 128 lanes of PCIe Gen4 per socket and support for 192 PCIe Gen4 lanes in 2P configuration that are bifurcatable down to x2, Ampere^® Altra^® provides maximum flexibility to interface with off-chip devices including networking cards up to 100 GbE or more and storage / NVMe devices, making it well suited for big data applications.

Ampere^® Altra^® supports cache coherent connectivity to off -chip accelerators. 64 lanes out of 128 PCIe Gen 4 lanes support CCIX, that could be used for networking, storage, or accelerator connectivity.

Ampere^® Altra^® provides industry leading power efficiency/core, while packing 80 cores in a single socket and 160 cores in a dual socket, establishing new levels of power efficiency with scalability.

Ampere’s power optimized design, coupled with 7nm process technology, enables the Ampere^® Altra™ processor to pack in more cores than any other data center class processor – all on a single die –  enabling datacenter infrastructure providers more cores per rack.

Ampere^® Altra^® processor’s advanced power management capabilities include Advanced Configuration Power Interface (ACPI) v6.2 support., Dynamic Frequency Scaling (DFS), on-die thermal monitoring, dynamic power estimation and Turbo mode support.