4. About the system
With G-Core Labs Cloud, you can access IPU-based hardware that has been configured with the necessary software. This means you can easily run your programs created in standard machine-learning frameworks like TensorFlow and PyTorch.
This section describes the hardware you will have access to and how it is configured. In addition, this section describes the software resources that are available, including the storage available.
You will be allocated your system as an entity called a “virtual Pod” (vPod). The vPod is provisioned as a single Virtual-IPU partition. A Virtual-IPU partition represents a number of IPUs which can communicate with one another. They are isolated so that all communication from physically neighbouring devices that are not in the same partition is prohibited.
You will have access to:
A secure vPod, fully isolated by a VLAN, running Ubuntu 20.04 or Ubuntu 18.04.
Server(s) (the number of servers you have access to depends on the Pod you have chosen. A server runs the Poplar graph engine and communicates with IPUs.)
V-IPU client software.
Local storage (exact amount depends on the region selected); this will be shared by all users accessing your vPod.
Root access is available via
Docker is pre-installed.
Full pricing details are available from G-Core Labs.
Your vPod instance will be set up with the latest released versions of the IPU-M software and Poplar SDK. If new versions are released later you will need to request an upgrade.
You access the vPod via an SSH connection. You can add multiple users to your vPod instance.
Full details are given in the Section 6, Access.
Multiple user jobs can be run at the same time provided that the total number of IPUs requested does not exceed the number of IPUs in the vPod. This is because we use a reconfigurable Virtual-IPU partition that supports many smaller partition requests within it. Refer to the V-IPU User Guide for more information on partitions.
4.3. Installed software
There are several software packages already installed on your vPod instance including:
Python (3.6 on Ubuntu 18.04 and 3.8 on Ubuntu 20.04)
Graphcore Poplar SDK (in
/opt/gc), which includes:
the Poplar Graph Programming Framework
PyTorch for the IPU
TensorFlow 1 and TensorFlow 2 for the IPU
Your vPod instance will always be set up with the latest released version of the Poplar SDK. Thereafter, you need to request an upgrade of this software.
Graphcore tutorials and examples
These repositories have been cloned into the
graphcoredirectory that exists in the default working directory when logging in.
4.4. Using your own data
You can upload your own data, within the available disk space.
4.5. Using synthetic data
You can use synthetic data to test model performance without the overhead of data transfer. The documentation for each framework describes how to enable using synthetic data, for example: