Barrel processor

From Wikipedia, the free encyclopedia

A barrel processor is a CPU that switches between threads of execution on every cycle. This CPU design technique is also known as "interleaved" or "fine-grained" multithreading.

For example, certain CDC Cyber computers executed one instruction from each of 20 different threads before returning to the first thread. Also, the IP3023 processor from Ubicom executes one instruction from each of 8 different threads before returning to the first thread.

Like preemptive multitasking, each thread of execution is assigned its own program counter and other registers (each thread's architectural state). However, a barrel processor can guarantee that each thread will get to execute 1 instruction every N cycles. In contrast, a preemptive multitasking machine typically runs one thread of execution for hundreds or thousands of cycles while all other threads wait their turn.

A technique called C-slowing can take a normal single-tasking processor design and automatically generate a corresponding barrel processor design. An n-way barrel processor generated this way acts much like n separate multiprocessing copies of the original single-tasking processor, each one running at roughly 1/n the original speed.

1 Advantages
2 Disadvantages
3 See also
4 External links

[edit] Advantages

A single-tasking processor spends a lot of time waiting around, not doing anything useful, whenever a cache miss or pipeline stall occurs. Advantages to employing barrel processors over single-tasking processors include:

The ability do useful work on the other threads while the stalled thread is waiting.
Designing an n-way barrel processor with n-deep pipelines is much simpler than designing a single-tasking processor because a barrel processor never has a pipeline stall and doesn't need feed-forward circuits.

Barrel processors have many advantages over single-tasking processors in real-time applications, because they can guarantee that a "real-time" thread can execute with precise timing, no matter what happens to the other threads -- even if some other thread locks up in an infinite loop or is continuously interrupted by hardware interrupts.

[edit] Disadvantages

There are, however, some disadvantages to barrel processors. All the threads must share the same cache, which can slow overall performance. This can especially slow down tasks which require low latency and depend heavily on cache memory. In addition, if there are insufficient threads available to run, a barrel processor may not have anything useful to do during a stall.