We have heard of CPUs and GPUs, now, NVIDIA with the help of its recent acquisition Mellanox is bringing a new class of processors to power up deep learning applications — DPUs or data processing units.
What is DPU?
DPUs or Data Processing Units, originally popularized by Mellanox, now wear a new look with NVIDIA; Mellanox was acquired by NVIDIA earlier this year. DPUs are a new class of programmable processor that consists of flexible and programmable acceleration engines which improve applications performance for AI and machine learning, security, telecommunications, storage, among others.
The team at Mellanox has already deployed the first generation of Bluefield DPUs in leading high-performance computing, deep learning, and cloud data centers to provide new levels of performance, scale, and efficiency with improved operational agility.
A DPU based SmartNIC is a network interface card that offloads processing tasks that the system CPU would normally handle. Using its own on-board processor, the DPU based SmartNIC may be able to perform any combination of encryption/decryption, firewall, TCP/IP and HTTP processing.
Why Do We Need DPU?
These DPUs are known by the name of Bluefield that has a unique design that can enable programmability to run at speeds of up to 200Gb/s. The Bluefield DPU integrates the NVIDIA Mellanox Connect best-in-class network adapter, encompassing hardware accelerators with advanced software programmability to deliver diverse software-defined solutions.
Organisations that rely on cloud-based solutions, especially can benefit immensely from DPUs. Here are few such instances, where DPUs flourish: –
- Enabling storage, networking, and security to be part of the composable infrastructure for cloud service providers
- Allowing the operator to facilitate bare metal to the cloud tenant as a service while preserving control over the server and protecting the environment
- can enable and isolate environment to accelerate compute-intensive security functions
- Can store, compute and secure data at the highest speeds while lowering cost and time by analyzing data at the edge
As of late last year, the team at Mellanox stated that they are actively researching into various platforms and integrating schemes to leverage the cutting-edge acceleration engines in the DPU-based SmartNICs for securing cloud-native workloads at 100Gb/s.
According to NVIDIA, a DPU comes with the following features:
- Data packet parsing, matching, and manipulation
- GPU-Direct accelerators to bypass the CPU and feed networked data directly to GPUs
- Traffic shaping “packet pacing” accelerator to enable streaming 4K/8K Video
- Precision timing accelerators for 5G capabilities
- Crypto acceleration
- Virtualization support
- Secure Isolation