SLIDE 1
Multicore Processors
Big deal? or No big deal?
1
Steven Parker SCI Institute School of Computing University of Utah
Multicore Processors Big deal? or No big deal? Steven Parker SCI - - PowerPoint PPT Presentation
Multicore Processors Big deal? or No big deal? Steven Parker SCI - - PowerPoint PPT Presentation
Multicore Processors Big deal? or No big deal? Steven Parker SCI Institute School of Computing University of Utah 1 SCI Institute SCI Institute software 3 No big deal SMP machines have been available for decades ccNUMA for 15+ 4 No big
SLIDE 2
SLIDE 3
SCI Institute software
3
SLIDE 4
No big deal
SMP machines have been available for decades ccNUMA for 15+
4
SLIDE 5
No big deal
SMP machines have been available for decades ccNUMA for 15+ Multi-core CPUs look like slightly crippled versions of the above
5
SLIDE 6
No big deal
SMP machines have been available for decades ccNUMA for 15+ Multi-core CPUs look like slightly crippled versions of the above Dual core laptop equivalent to high-end 1990 Workstation My desktop is equivalent to small SGI Challenge
6
SLIDE 7
SLIDE 8
C-SAFE Computations
8
Hundreds to thousands of CPUs Hundreds of wall-clock hours Millions of hours of CPU time per run Dozens of runs for a study All can handle increase in parallelism immediately
SLIDE 9
Big deal?
Cost for 2-core machine
1990: $10000 2007: $1000
Cost for 8-core machine
1990: $150k 2007: $6k
Big deal: much more available
9
SLIDE 10
Parallelism opportunities
SIMD instructions available for ~decade 4-way parallelism broadly available Very little use (mainly libraries) Difficulty of programming a challenge What it it became 8-way or 16-way or 128 way? Lesson: use it or lose it
10
SLIDE 11
Programming Models
Straight C code inadequate How to efficiently handle:
Graph evaluations (updates) High load imbalance Lazy evaluation
Reliability in a complex system elusive Parallelism not composable Big deal: revitalization in parallel computing research?
11
SLIDE 12
Scalability
Multi-core means a 1000 CPU cluster now has 10000 CPUs
10x increase in impact of serial code 90% efficient means < 1% of code is serial/duplicated What about 100k cpus?
Opportunity: programming models that allow multiple levels of parallelism
More than MPI + OpenMP More than threads Big deal: opportunities for research in programming models, libraries, systems
12
SLIDE 13
Education
95% of CS grad students think they can program
13
* 80% of statistics are made up on the spot, including these
SLIDE 14
Education
95% of CS grad students think they can program 50% actually can
14
* 80% of statistics are made up on the spot, including these
SLIDE 15
Education
95% of CS grad students think they can program 50% actually can 50% of CS grad students think they can write a multithreaded program
15
* 80% of statistics are made up on the spot, including these
SLIDE 16
Education
95% of CS grad students think they can program 50% actually can 50% of CS grad students think they can write a multithreaded program 10% actually can
16
* 80% of statistics are made up on the spot, including these
SLIDE 17
Education
95% of CS grad students think they can program 50% actually can 50% of CS grad students think they can write a multithreaded program 10% actually can 1% can make it efficient and maintainable
17
* 80% of statistics are made up on the spot, including these
SLIDE 18
Education
95% of CS grad students think they can program 50% actually can 50% of CS grad students think they can write a multithreaded program 10% actually can 1% can make it efficient and maintainable
17
Should we teach P(n) instead of O(n)?
* 80% of statistics are made up on the spot, including these
SLIDE 19
Education
95% of CS grad students think they can program 50% actually can 50% of CS grad students think they can write a multithreaded program 10% actually can 1% can make it efficient and maintainable
17
Should we teach P(n) instead of O(n)? EAE section of 2420 will!
* 80% of statistics are made up on the spot, including these
SLIDE 20
Observations/Thoughts?
Memory latency (in cycles) no longer going up Memory bandwidth per core going down
Locality aware algorithms are still important
Speculative execution may go away for some things Terminology outdated: what is a CPU? Ask an oldtimer what “core” means APIs outdated (locality control) Is the # of cores as useless as Mhz/Ghz was?
18
SLIDE 21
Economics
Q: How many cores will a 2020 CPU contain? A: As many as we can convince people to pay for. What are those applications? What are those algorithms? What are those systems?
19
SLIDE 22
More questions than answers
20