![The
Manycore
Problem
(from
[2])
• Shared
memory
10
words
access
very
expensive
1
word
0
250
500
750
1000
– Inter-‐core
messaging
DMA
Load
Mem.
– Purpose-‐specific
networks
Msg:
3
hops
Msg:
1
hop
23/08/11
10](https://image.slidesharecdn.com/imec-sly-and-roarvm-flattened-110824075830-phpapp01/85/Sly-and-the-RoarVM-Exploring-the-Manycore-Future-of-Programming-10-320.jpg)
![Read-‐Mostly
Heap
[2]
• TILE64
cache
coherency
VERY
expensive
– Read-‐mostly
heap:
incoherent,
‘user
managed
CC’
Cache
Cache
Cache
Read/Write
Read-‐mostly
23/08/11
15](https://image.slidesharecdn.com/imec-sly-and-roarvm-flattened-110824075830-phpapp01/85/Sly-and-the-RoarVM-Exploring-the-Manycore-Future-of-Programming-15-320.jpg)
![Pauseless
GC
• Based
on
a
paper
by
Click
et
al.,
2005
[3]
• No
global
synchronizaWon
– Local
checkpoints:
1
mutator
and
GC
thread
• ConWnuously
running
GC
threads
• Load-‐Value-‐barrier
+
self-‐healing
of
stale
refs.
• Constant
performance
overhead
– Without
hardware
support
• Easy
to
implement
(2
MA
students,
1
month)
23/08/11
18](https://image.slidesharecdn.com/imec-sly-and-roarvm-flattened-110824075830-phpapp01/85/Sly-and-the-RoarVM-Exploring-the-Manycore-Future-of-Programming-18-320.jpg)
![References
and
Material
[1]
J.
Pallas
and
D.
Ungar.
MulWprocessor
Smalltalk:
A
Case
Study
of
a
MulWprocessor-‐based
Programming
Environment.
In
PLDI
’88,
p.
268–277.
ACM,
1988.
[2]
D.
Ungar
and
S.
Adams.
HosWng
an
Object
Heap
on
Manycore
Hardware:
An
ExploraWon.
In
DLS’09,
p.
99-‐110.
ACM,
2009.
[3]
C.
Click,
G.
Tene,
and
M.
Wolf.
The
Pauseless
GC
Algorithm.
In
VEE
’05,
p.
46-‐56.
ACM,
2005.
[4]
D.
Ungar,
D.
Kimelman,
and
S.
Adams.
Inconsistency
Robustness
for
Scalability
in
InteracWve
Concurrent‑Update
In-‐Memory
MOLAP
Cubes.
In
Inconsistency
Robustness
Symposium
2011.
[5]
P.
Inostroza
Valdera.
The
Sly3
Programming
Language:
An
IntroducWon
and
Analysis.
h=p://so@.vub.ac.be/~smarr/renaissance/sly3-‐overview/
[6]
Renaissance
Project:
h=p://so@.vub.ac.be/~smarr/renaissance/
23/08/11
Roundup
and
Conclusion
22](https://image.slidesharecdn.com/imec-sly-and-roarvm-flattened-110824075830-phpapp01/85/Sly-and-the-RoarVM-Exploring-the-Manycore-Future-of-Programming-22-320.jpg)
Sly and the RoarVM: Exploring the Manycore Future of Programming
- 1. Sly and the RoarVM Exploring the Manycore Future of Programming Renaissance Project h=p://[email protected]/~smarr/renaissance/ Stefan Marr So@ware Languages Lab, Vrije Universiteit Brussel ExaScience Lab, 2011-‐08-‐24
- 2. Agenda • Context: Renaissance Project • Sly – Demo – Language Overview • RoarVM – Intercore Messaging – Read-‐mostly heap – Pauseless GC • Outlook 23/08/11 2
- 3. Renaissance Project • CollaboraWon between – IBM Research, Portland State Univ, VUB • Problem – Parallel systems are nondeterminisWc – CommunicaWon is expensive • The vision – Embrace nondeterminism – Harness emergence – Confidence → Delay 23/08/11 3
- 4. Demo 24/08/11 4
- 5. Possible Real Life ApplicaWon Image: h=p://blogs.technet.com/b/andrew/ archive/2007/08/22/olap-‐cubes-‐and-‐ mulWdimensional-‐analysis.aspx 23/08/11 5
- 6. Possible Real Life ApplicaWon • ACK: ongoing work at IBM Research • Online analyWcal processing (OLAP) – Business intelligence, reporWng, data mining • Query mulWdimensional data sets – Includes dependent/calculated data dimensions – Sync. required to avoid use of stale data • Without sync.: how to bound error? – RecalculaWng, refreshing results 23/08/11 6
- 7. Sly • Ensembles: collecWons represenWng a whole, mulW-‐part enWWes – e.g. a flock of birds • Adverbs: modifiers to operands – individualLY, serialLY, randomLY: n • Gerunds: reducWon semanWcs – averagING, selectINGpredicate, ensemblING 23/08/11 7
- 8. Examples of Boids • Every boid is its own thread, no synch. computeCentroid ^ self flock boids averagINGserialLYposition computeNeighbors ^ self flock boids selectINGisNear: self matchVelocityWithNeighbors ^ self neighbors averagINGvelocity / 8.0 More details: h=p://[email protected]/~smarr/renaissance/sly3-‐overview/ 23/08/11 8
- 9. Virtual Machines for Manycore Architectures ROARVM 23/08/11 9
- 10. The Manycore Problem (from [2]) • Shared memory 10 words access very expensive 1 word 0 250 500 750 1000 – Inter-‐core messaging DMA Load Mem. – Purpose-‐specific networks Msg: 3 hops Msg: 1 hop 23/08/11 10
- 11. Architecture Overview • Core = Interpreter Instance = OS Process/Thread • ‘Single System Image’/shared memory VM 23/08/11 11
- 12. Message-‐Passing between Interpreters • To avoid main-‐memory latency • VM internal use only – GC • preGCAcWon + ACK • doAllRootsHere + Response, … – SafepoinWng • requestSafepointOnOtherCores • grantSafepoint, … 23/08/11 12
- 13. Memory System • Object table, moving stop-‐the-‐world GC – Usually allocaWon on heap of local core Object Heap 24/08/11 13
- 14. Memory System • Object table, moving stop-‐the-‐world GC – Usually allocaWon on heap of local core Global Safepoint Object Heap 23/08/11 14
- 15. Read-‐Mostly Heap [2] • TILE64 cache coherency VERY expensive – Read-‐mostly heap: incoherent, ‘user managed CC’ Cache Cache Cache Read/Write Read-‐mostly 23/08/11 15
- 16. Stop-‐the-‐World in Manycore Era? Work Run out of space Arrival at safe point Do garbage collect Work 23/08/11 16
- 17. Stop-‐the-‐World in Manycore Era? … … … … 23/08/11 17
- 18. Pauseless GC • Based on a paper by Click et al., 2005 [3] • No global synchronizaWon – Local checkpoints: 1 mutator and GC thread • ConWnuously running GC threads • Load-‐Value-‐barrier + self-‐healing of stale refs. • Constant performance overhead – Without hardware support • Easy to implement (2 MA students, 1 month) 23/08/11 18
- 19. OUTLOOK 23/08/11 19
- 20. Conclusion • Sly: map/reduce-‐like programming model – Based on ensembles, adverbs and gerunds • RoarVM – Research arWfact, stable, used daily – No opWmizaWon of sequenWal performance – Tested up to 59cores – Shows weak scalability 23/08/11 20
- 21. Outlook • Sly – Ideas for non-‐linear syntax – Algorithms for nondeterminisWc programming • RoarVM: focus on research – VMs are mulW-‐language plarorms – Support for concurrency models – DSLs for specific concurrent/parallel problems – Language guarantees in inter-‐model interacWon 23/08/11 21
- 22. References and Material [1] J. Pallas and D. Ungar. MulWprocessor Smalltalk: A Case Study of a MulWprocessor-‐based Programming Environment. In PLDI ’88, p. 268–277. ACM, 1988. [2] D. Ungar and S. Adams. HosWng an Object Heap on Manycore Hardware: An ExploraWon. In DLS’09, p. 99-‐110. ACM, 2009. [3] C. Click, G. Tene, and M. Wolf. The Pauseless GC Algorithm. In VEE ’05, p. 46-‐56. ACM, 2005. [4] D. Ungar, D. Kimelman, and S. Adams. Inconsistency Robustness for Scalability in InteracWve Concurrent‑Update In-‐Memory MOLAP Cubes. In Inconsistency Robustness Symposium 2011. [5] P. Inostroza Valdera. The Sly3 Programming Language: An IntroducWon and Analysis. h=p://[email protected]/~smarr/renaissance/sly3-‐overview/ [6] Renaissance Project: h=p://[email protected]/~smarr/renaissance/ 23/08/11 Roundup and Conclusion 22