Many-Core Programming with GRAMPS

1
Many-Core Programming with GRAMPS

• Jeremy Sugerman
• Kayvon Fatahalian
• Solomon Boulos
• Kurt Akeley
• Pat Hanrahan

2
Problem Statement

• Facilitate efficient development and execution in
  many-/multi-core commodity systems.
• Homogeneous or heterogeneous cores.
• Status Quo
• GPUs Easy to write GL/D3D and run it fast, hard
  to express anything else
• CPUs Possible (not easy) to write anything,
  possible (hard) to run it fast

3
GRAMPS Background

• Resembles a GPU with software constructed
  pipeline.
• Not (too) radical even in a pure graphics context
• Similar story saw fixed -gt programmable shading
• Now the pipeline topology is under analogous
  pressures proliferation of stages and options
• And graphics is more than a GL/D3D pipeline
• And throughput / many-core is more than graphics

4
GRAMPS Programming Model

• Software constructs the pipeline (actually graph)
• Exposes threads, shaders, fixed function stages
• Coprocessors exposed via ISA
• Exposes FIFOs / Queues connecting stages
• Also enables software push / re-sorting
• Exposes Buffers for memory access

5
GRAMPS Place

• Compared to GPU Pipeline
• More things possible (and medium easy), still
  (mostly) runs fast, less hardware independent
• Compared to CPU
• Easier to write things, easier to run them well,
  some loss of expressivity and flexibility
• Still a role for a graphics pipeline. Its an
  app!
• GRAMPS is a layer, model for state machines.

6
GRAMPS and Streaming

• From some angles, GRAMPS sounds a lot like Stream
  Processing / Computing
• Distinctions are most visible in the target
  traits.
• Streaming expects predictable data creation,
  flow, and consumption. Intensive offline /
  compile-time optimization and pre-scheduling.
• GRAMPS expects dynamic data-dependent execution,
  (and thus) run-time scheduling
• Also, GRAMPS assumes commodity and heterogeneity.

7
GRAMPS Examples
Rasterization Pipeline
Input Fragment Queue
Output Fragment Queue
Frame Buffer
Shade
FB Blend
Rast
Ray Queue
Ray Tracing Pipeline
Intersect
Camera
Sample Queue
Frame Buffer
Pixel Queue
FB Blend
Shade
8
GRAMPS Overview

• Concepts
• Graphs
• Stages thread, shader, fixed-function
• Queues ordered, unordered, sets (exclusion)
• Buffers
• Components
• APIs setup/driver, thread, shader
• Scheduler fat core, shader core, top-level

9
What Weve Built

• Three rendering pipelines
• Direct3D, Packet Tracer, D3D Push (Hybrid)
• Simulator and Runtime for two machines
• GPU-like Many threads per core, hw sched
• CPU-like Few threads per core, sw sched

10
Rendering Pipelines
11
Initial Results

• Measured thread occupancy, worst case total queue
  memory.

12
GRAMPS Vis
13
High-level Challenges

• Is GRAMPS a suitable GPU evolution?
• Enable pipeline competitive with bare metal?
• Enable innovation advanced / alternative
  methods?
• Is there a best graphics pipeline on top?
• Is GRAMPS a good parallel compute model?
• Map well to hardware, hardware trends?
• Support important apps?
• Concepts influence developers?

14
Whats Next?

• Low level implementation scheduling, more
  accurate simulation.
• More apps REYES, physics, likely more.
• Audit and refine model graph modification /
  state change, fork-join / blocking calls, locks /
  barriers / synchronization primitives intra- or
  inter-stage
• Prototype, explore next generation graphics
  pipelines.