Hardware device comprising multiple accelerators for performing multiple independent hardware acceleration operations

A circuit card assembly configured to provide hardware acceleration for a software process running on a computing device, the computing device in communication with the circuit card assembly over a PCI bus, the assembly comprising:

• a first hardware accelerator configured to perform only a decompression operation;
• a second hardware accelerator configured to perform only a virus detection operation;
• a memory on the accelerator card assembly configured to provide data between at least the first accelerator and the second accelerator;
• and
• means for receiving, from the computing device over the PCI bus, a data structure indicating one or more operations to be performed by each of the first and second hardware accelerators of the circuit card assembly and based thereon, the means separate and independent from the computing device and each accelerator, scheduling an operation of the first hardware accelerator and scheduling an operation of the second hardware accelerator, including originally generating headers initiating respective operations independently of the computing device, wherein
• the first hardware accelerator is configured to, after the operation of the first hardware accelerator is completed, i) store output data of the first hardware accelerator in the memory on the accelerator card assembly, and ii) send a synchronization message to the second hardware accelerator to perform the operation of the second hardware accelerator, wherein output data from the first hardware accelerator is used as input data for the second hardware accelerator.

A computing system comprising:

• a bus for transmitting data between a plurality of devices in communication with the bus;
• a central processing unit in communication with the bus;
• a memory device in communication with the bus;
• a plurality of hardware accelerators configured to perform algorithmic operations on data;
• including a first hardware accelerator configured to perform only a decompression operation and a second hardware accelerator configured to perform only a virus detection operation;
• a task management unit separate and independent from the central processing unit and each hardware accelerator, the task management unit in communication with each of the plurality of hardware accelerators and the central processing unit, wherein the central processing unit transmits to the task management unit a data structure including a plurality of identifiers, each of the identifiers identifying one of the plurality of hardware accelerators, and the task management unit generates headers for transmission to respective hardware accelerators in order to initiate execution of the hardware accelerator operations identified in the data structure without further input from the central processing unit,
• wherein each of the plurality of hardware accelerators is configured to, after the operation of a first one of the plurality of hardware accelerators is completed, i) store output data of the first hardware accelerator in the memory device, and ii) send a synchronization message to a next one, determined by the header generated by the task management unit, of the plurality of hardware accelerators to perform a next operation, wherein output data from the first hardware accelerator is used as input data for the next hardware accelerator.