1. Exascale Programming Models Lecture Series 06/12/2014 What is OCR? TG Team (presenter: Romain Cledat) June 12, 2014 https://xstackwiki.modelado.org/Traleika_Glacier/ This research was, in part, funded by the U.S. Government, DOE and DARPA. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Government.

2. Exascale Programming Models Lecture Series 06/12/2014 OCR – Open Community Runtime – Developed collaboratively with partners (mainly Rice University and Reservoir Labs) The term ‘OCR’ is used to refer to way too many concepts – A programming model – A user-level API – A runtime framework – One of a multitude of reference runtime implementationsOCR 2

3. Exascale Programming Models Lecture Series 06/12/2014 Design a software stack to meet Exascale goals – Target a strawman architecture – Provide a programming model, API, reference implementation and tools Concerns – Extreme hardware parallelism – Data locality – Fine grained resource management – Resiliency – Power and energy and not just performance – Platform independence TG X-Stack project goals 3

4. Exascale Programming Models Lecture Series 06/12/2014 mainEdt fibIterEdt sumEdt N finishEdt N-2 N-1 Dataflow programming model 4 EDT Datablock Create Event Runtime maps the constructed data-flow graph to architecture ……….. Shared LLC Interconnect ………..

5. Exascale Programming Models Lecture Series 06/12/2014 OCR level of abstraction 5 void ParallelAverage( float* output, const float* input, size_t n ) { Average avg; avg.input = input; avg.output = output; parallel_for( blocked_range ( 1, n ), avg ); } if(!range.empty()) { start_for& a = *new(task::allocate_root()) start_for(range,body,partitioner); task::spawn_root_and_wait(a); } void generic_scheduler::local_spawn_root_and_wait( task& first, task*& next ) { internal::reference_count n = 0; for( task* t=&first; ; t=t->prefix().next ) { ++n; t->prefix().parent = &dummy; if( &t->prefix().next==&next ) break; } dummy.prefix().ref_count = n+1; if( n>1 ) local_spawn( *first.prefix().next, next ); local_wait_for_all( dummy, &first ); } hides… OCR’s level of abstraction is at the very bottom TBB user-friendly API

6. Exascale Programming Models Lecture Series 06/12/2014 Common – All objects globally and uniquely identifiable and relocate-able Computation – Event Driven Task (EDT) – Does not perform synchronization – Distinct from the notion of thread or core Data – Data-block (DB) – Relocate-able consecutive chunk of data Synchronization, links – Events – Runtime-visible Slots – Positional end-points for dependences OCR concepts 6

7. Exascale Programming Models Lecture Series 06/12/2014 N pre slots (N known at creation time) Optional attached “completion event” OCR concepts: building blocks 7 Evt 0N EDT 0N ( ) Data No pre slots Post slot always “satisfied” N pre slots (N fixed by type of event NOT determined by user) Post slot initially “unsatisfied” Slot is: – Connected (attached to another slot) or unconnected – Satisfied (user-triggered or runtime-triggered) or unsatisfied Pre slots Post slots (multiple connections)

8. Exascale Programming Models Lecture Series 06/12/2014 OCR concepts: add dependence 8 Data Evt 0N OR EDT 0N Evt 0N OR Evt 0N EDT 0N Connected => 1 of 4 possible combinations Argument 1 Argument 2

9. Exascale Programming Models Lecture Series 06/12/2014 OCR concepts: satisfy 9 EDT 0N Evt 0N OR Data OR NULL EDT 0N Satisfied/triggered Data => 1 of 4 possible combinations Argument 1 Argument 2

10. Exascale Programming Models Lecture Series 06/12/2014 EDTs – 0..N in/out pre-slots Slots are initially “unconnected” and “unsatisfied” At creation time, the number of incoming slots must be known – An EDT executes after all pre slots are “satisfied” Satisfaction of pre slots can happen in any order – An EDT can access memory: Data-blocks: – passed in through one of its in/out slots (the EDT gets a C pointer) – created by the EDT Stack and ephemeral heap (local) NO global memory – An EDT, during its execution, can at any time: Write to any accessible data-blocks Manipulate the dependence graph for future (not yet runnable) EDTs by adding dependences, satisfying events, etc. OCR execution model for EDTs 10

11. Exascale Programming Models Lecture Series 06/12/2014 Dynamic dependence construction Producer and consumer never know about each other Focus on minimum needed for placement and scheduling Example 1: Producer/Consumer 11 Consumer EDT Producer EDT Data ConceptOCR Evt Consumer EDT Producer EDT Data (1) dbCreate (*) addDep (3) satisfy (2) edit Data Who executes call Data dependence Control dependence

12. Exascale Programming Models Lecture Series 06/12/2014 Control dependence is no different than a data dependence Example 2: Simple synchronization 12 (1) satisfy ConceptOCR Step 1 EDT Step 2-a EDT Step 2-b EDT Evt Step 1 EDT (*) addDep NULL Step 2-a EDT Step 2-b EDT

13. Exascale Programming Models Lecture Series 06/12/2014 Events – 0..N pre slots Slots are initially “unconnected” and “unsatisfied” – Events have a “trigger” rule that determines when their post slot transitions to “satisfied” and what gets connected to it Simple event (pass-through) – 1 pre slot – When: satisfy post slot on incoming slot satisfaction – What: whatever is on incoming slot (pass GUID) Latch event (multi-party synchronization) – 2 pre slots; “waiting-on” count and current count – When: satisfy outgoing slot when number of satisfies on both pre slots matches (similar to reference count in TBB) – What: NULL (incoming data-blocks are ignored) OCR execution model for events 13

14. Exascale Programming Models Lecture Series 06/12/2014 Example 3: In place parallel update 14 ConceptOCR Setup EDT Parallel_1 EDT Parallel_2 EDT Wrapup EDT Data Setup EDT Data Parallel_1 EDT Parallel_2 EDT Finish EDT Wrapup EDT (1) dbCreate (1) edtCreate (3) edtCreate (4) addDep (2) addDep (3) edtCreate

15. Exascale Programming Models Lecture Series 06/12/2014 Example 4: Single assignment update 15 Concept OCR Setup EDT Parallel_1 EDT Parallel_2 EDT Wrapup EDT Data Setup EDT Data Parallel_1 EDT Parallel_2 EDT Wrapup EDT (1) dbCreate (1) edtCreate (2) addDep Data2Data1 Evt2 Data2Data1 Evt1 (4) dbCreate (5) satisfy (3) addDep (1) evtCreate

16. Exascale Programming Models Lecture Series 06/12/2014 OCR ecosystem FSim - TG Architecture Low-level compilers Platforms OCR implementations LLVM OCR targeting TG C, Array DSL CnC Hero Code HC CnC Translator HC Compiler R-Stream HTA PIL Programming platforms OCR API + Tuning Annotations Open Community Runtime x86 GCC OCR targeting x86 Cluster Evaluation platforms

17. Exascale Programming Models Lecture Series 06/12/2014 OCR API is at the “assembly” level; other tools are meant to sit between it and programmers Few simple concepts, multiple ways to use them – Interested in determining “best” use Dependence graph built on the fly: – Complicates the writing of the program – Scalable approachTake-aways 17

18. Exascale Programming Models Lecture Series 06/12/2014 On some code, OCR matches or bests OMP Simple scheduler, no data-blocks (very preliminary but promising) Preliminary results 18

19. Exascale Programming Models Lecture Series 06/12/2014 Development of a specification: – Memory model Tuning hints and annotations More expressive support for collectives Areas of investigation 19

20. Exascale Programming Models Lecture Series 06/12/2014 20 Backup

21. Exascale Programming Models Lecture Series 06/12/2014 Strawman architecture 21 Intel Confidential / Internal Use Only Heterogeneous Hierarchical architecture Tapered memory bandwidth Global, shared address space Software managed non- coherent memories Functional simulator available DP FP FMAC DP FP FMAC Execution Engine (XE) 32KB I$ 64KB SP RF? Application specific GP Int GP Int Control Engine (CE) 32KB I$ 64KB SP RF? System SW XE CE 1MB shared L2 Block (8 XE + CE) Cluster (16 Blocks) ……….. 8MB Shared LLC Interconnect ……….. Processor Chip (16 Clusters)

22. Exascale Programming Models Lecture Series 06/12/2014 OCR vs other solutions 22 CnCMPIOCROpenM P TBB Execution model TasksBulk SyncFine- grained tasks Bulk SyncTasks Memory model Shared memory Explicit message passing Explicit; global Shared memory Separation of concerns? YesNoYesNoYes (but can dig deeper)