SLIDE 1
Background
- Digital media processing has become
pervasive
- Real-time processing requires large
amounts of computation and bandwidth
- Over time, as resources increase,
Imagine: Media Processing with Streams Brucek Khailany et al. and - - PowerPoint PPT Presentation
Imagine: Media Processing with Streams Brucek Khailany et al. and - - PowerPoint PPT Presentation
Imagine: Media Processing with Streams Brucek Khailany et al. and a little bit of Evaluating the Imagine Stream Architecture Jung Ho Ahn et al. Presented by Dan Amelang Background Digital media processing has become pervasive
SLIDE 2
SLIDE 3
Stream Processing
- Good fit for media processing
- Computationally intensive
- Highly parallel and independent data
- High latency tolerance
- Little data reuse
- Simple control
- Communication and parallelism explicit
SLIDE 4
Architecture Options
- General purpose architecture
– Caches optimized for latency and data reuse – Don't provide enough functional units – Large multiported register file inefficient
- ASIC
– Efficient and fast – Limited use
- Stream Processor
– Trade-off between programmability and efficiency
SLIDE 5
Streams and Kernels
SLIDE 6
Imagine
SLIDE 7
Programming Model
- StreamC for stream and kernel interaction
- KernelC for VLIW kernel code
SLIDE 8
Memory System
- DRAM <-> SRF controlled by host
- SRF <-> LRF at the request of the kernel
- LRF <-> LRF statically scheduled by the
compiler
- Streams are composed of 32 word blocks
- SRF transfers go through stream buffers of
2 blocks
SLIDE 9
6 Arithmetic Clusters
SLIDE 10
Simulation vs. Prototype
- 500 MHz vs. 200 MHz
- 20 GFLOPS vs. 8 GFLOPS
- Cut bandwidths in half
- Double power consumption
- Halve performance
SLIDE 11
SLIDE 12
Kernel Performance Breakdown
SLIDE 13
Application Performance Breakdown
SLIDE 14
"Where are they now?"
- Imagine became basis of new company
"Stream Processors, Inc"
- Merrimac
- StreamIt
- BrookGPU
- Last month, Bill Dally was appointed VP of