DOCSLIB.ORG

“Scalable” Cache Coherence Scalable Cache Coherence Directory Coherence Protocols MSI Directory Protocol MSI Directory Proto

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
“Scalable” Cache Coherence Scalable Cache Coherence Directory Coherence Protocols MSI Directory Protocol MSI Directory Proto “Scalable” Cache Coherence Scalable Cache Coherence LdM/StM CPU($) CPU($) Mem R R Mem I Mem R R Mem CPU($) CPU($) Part I: bus bandwidth • Point-to-point interconnects Replace non-scalable bandwidth substrate (bus)… • Glueless MP: no need for additional “glue” chips …with scalable one (point-to-point network, e.g., mesh) + Can be arbitrarily large: 1000’s of processors • Massively parallel processors (MPPs) • Only government (DoD) has cache-coherent MPPs… Part II: processor snooping bandwidth • Companies have much smaller systems: 32–64 processors • Scalable multi-processors • Most snoops result in no action • AMD Opteron/Phenom – point-to-point, glueless, broadcast • Replace non-scalable broadcast protocol (spam everyone)… • Distributed memory: non-uniform memory architecture (NUMA) …with scalable directory protocol (only notify processors that care) • Multicore: on-chip mesh interconnection networks 57 58 57 58 Directory Coherence Protocols MSI Directory Protocol Observe: address space statically partitioned LdMiss, • Processor side + Can easily determine which memory module holds a given line StMiss • Directory follows its own protocol (obvious • That memory module sometimes called “home” I in principle) – Can’t easily determine which processors have line in their caches • Similar to bus-based MSI • Bus-based protocol: broadcast events to all processors/caches • Same three states ± Simple and fast, but non-scalable Store • Same five actions (keep BR/BW names) Directories: non-broadcast coherence protocol • Extend memory to track caching information • Minus red arcs/actions • For each physical cache line whose home this is, track: • Events that would not trigger action anyway • Owner: which processor has a dirty copy (i.e., M state) +Directory won’t bother you unless you need • Sharers: which processors have clean copies (i.e., S state) to act • Processor sends coherence event to home directory StMiss, WB Store • Home directory only sends events to processors that care M S • For multicore w/ shared L3, put directory info in cache tags LdMiss Load, Store Load, LdMiss 59 60 59 60 MSI Directory Protocol Directory Flip Side: Latency P0 P1 Directory Processor 0 Processor 1 • Directory protocols 0: addi r1,acctsr3 –:–:500 + Lower bandwidth consumption more scalable 1: ld 0(r3),r4 S:500 S:0:500 – Longer latencies 2 hop miss 3 hop miss 2: blt r4,r2,done 3: sub r4,r2r4 M:400 M:0:500 • Two read miss situations P0 P0 P1 4: st r4,0(r3) (stale) 0: addi r1,acctsr3 1: ld 0(r3),r4 S:400 S:400 S:0,1:400 • Unshared: get data from memory Dir Dir 2: blt r4,r2,done • Snooping: 2 hops (P0memoryP0) 3: sub r4,r2r4 • Directory: 2 hops (P0memoryP0) 4: st r4,0(r3) I: M:300 M:1:400 • Shared or exclusive: get data from other processor (P1) • Assume cache-to-cache transfer optimization ld by P1 sends BR to directory • Snooping: 2 hops (P0P1P0) • Directory sends BR to P0, P0 sends P1 data, does WB, goes to S st by P1 sends BW to directory – Directory: 3 hops (P0memoryP1P0) • Directory sends BW to P0, P0 goes to I • Common, many processors high probability someone has it 61 62 61 62 Directory Flip Side: Complexity Coherence on Real Machines • Latency is not the only issue for directories • Many uniprocessors designed with on-chip snooping logic • Subtle correctness issues as well • Can be easily combined to form multi-processors • Stem from unordered nature of underlying inter-connect • e.g., Intel Pentium4 Xeon • Individual requests to single cache must be ordered • And multicore, of course • Bus-based snooping: all processors see all requests in same order • Larger scale (directory) systems built from smaller MPs • Ordering automatic • e.g., Sun Wildfire, NUMA-Q, IBM Summit • Point-to-point network: requests may arrive in different orders • Directory has to enforce ordering explicitly • Some shared memory machines are not cache coherent • Cannot initiate actions on request B… • e.g., CRAY-T3D/E ..until all relevant processors complete actions on request A • Shared data is uncachable • Requires directory to collect acks, queue requests, etc. • If you want to cache shared data, copy it to private data • Directory protocols section • Obvious in principle • Basically, cache coherence implemented in software – Complicated in practice • Have to really know what you are doing as a programmer 63 64 63 64 Roadmap Checkpoint Tricky Shared Memory Examples • Answer the following questions: App App App • Thread-level parallelism (TLP) • Initially: all variables zero (that is, x is 0, y is 0, flag is 0, A is 0) System software • Shared memory model • What value pairs can be read by the two loads? (x, y) pairs: • Multiplexed uniprocessor thread 1 thread 2 Mem CPUCPU I/O CPUCPU • Hardware multithreading load x store 1 → y CPUCPU • Multiprocessing load y store 1 → x • Synchronization • (0,0) and (1,1) easy to see • Lock implementation • load x, store 1 → y, load y, store 1 → x gives (0,1) • Locking gotchas • Is it possible to get (1,0)? • Cache coherence • Bus-based protocols • Directory protocols • Memory consistency models 65 66 65 66 Tricky Shared Memory Examples Memory Consistency • Answer the following questions: • Memory coherence • Initially: all variables zero (that is, x is 0, y is 0, flag is 0, A is 0) • Creates globally uniform (consistent) view… • What value pairs can be read by the two loads? (x, y) pairs: …of a single memory location (in other words: cache line) thread 1 thread 2 – Not enough → load x store 1 y • Cache lines A and B can be individually consistent… load y store 1 → x …but inconsistent with respect to each other • What value pairs can be read by the two loads? (x, y) pairs: thread 1 thread 2 • Memory consistency store 1 → y store 1 → x load x load y • Creates globally uniform (consistent) view… • What value can be read by “Load A” below? …of all memory locations relative to each other thread 1 thread 2 • Who cares? Programmers store 1 → A while(flag == 0) { } store 1 → flag load A – Globally inconsistent memory creates mystifying behavior 67 68 67 68 Hiding Store Miss Latency Write Misses and Store Buffers Read miss? • Recall (back from caching unit) • Load can’t go on without the datamust stall • Hiding store miss latency Write miss? Processor Technically, no one needs datawhy stall? • How? Store buffer • SB Store buffer: a small buffer • Said it would complicate multiprocessors • Stores put addr/value to write buffer, keep going • Yes, it does! • Store buffer writes stores to D$ in the background Cache • Loads must search store buffer (in addition to D$) + Eliminates stalls on write misses (mostly) WBB – Creates some problems Next-level cache 69 70 69 70 Store Buffers & Consistency Coherence vs. Consistency A=flag=0; A=0 flag=0 Processor 0 Processor 1 Processor 0 Processor 1 A=1; while (!flag); // spin A=1; while (!flag); // spin flag=1; print A; flag=1; print A; • Intuition says: P1 prints A=1 Consider the following execution: • Coherence says: absolutely nothing • Processor 0’s write to A, misses the cache. Put in store buffer • P1 can see P0’s write of flag before write of A!!! How? • Processor 0 keeps going • P0 has a coalescing store buffer that reorders writes • Processor 0 write “1” to flag hits, completes • Or out-of-order execution • Processor 1 reads flag… sees the value “1” • Or compiler re-orders instructions • Processor 1 exits loop • Imagine trying to figure out why this code sometimes “works” and sometimes doesn’t • Processor 1 prints “0” for A • Real systems act in this strange manner Ramification: store buffers can cause “strange” behavior • What is allowed is defined as part of the ISA of the processor • How strange depends on lots of things 71 72 71 72 Memory Consistency Models Restoring Order • Sequential consistency (SC) (MIPS, PA-RISC) • Sometimes we need ordering (mostly we don’t) • Formal definition of memory view programmers expect • Prime example: ordering between “lock” and data • Processors see their own loads and stores in program order • How? insert Fences (memory barriers) + Provided naturally, even with out-of-order execution • Special instructions, part of ISA • But also: processors see others’ loads and stores in program order • Example • And finally: all processors see same global load/store ordering – Last two conditions not naturally enforced by coherence • Ensure that loads/stores don’t cross lock acquire/release operation • Corresponds to some sequential interleaving of uniprocessor orders acquire • Indistinguishable from multi-programmed uni-processor fence • Processor consistency (PC) (x86, SPARC) critical section • Allows a in-order store buffer fence • Stores can be deferred, but must be put into the cache in order release • Release consistency (RC) (ARM, Itanium, PowerPC) • How do fences work? • Allows an un-ordered store buffer • They stall execution until write buffers are empty • Stores can be put into cache in any order • Makes lock acquisition and release slow(er) • Use synchronization library, don’t write your own 73 74 73 74 Shared Memory Summary Flynn’s Taxonomy • Synchronization: regulated access to shared data • Proposed by Michael Flynn in 1966 • Key feature: atomic lock acquisition operation (e.g., t&s) • SISD – single instruction, single data • Performance optimizations: test-and-test-and-set, queue • Traditional uniprocessor locks • SIMD – single instruction, multiple data • Coherence: consistent view of individual cache lines • Execute the same instruction on many data elements • Absolute coherence not needed, relative coherence OK • Vector machines, graphics engines • VI
Load more

Recommended publications
  • Comparative Performance Evaluation of Cache-Coherent NUMA and COMA Architectures Abstract 1 Introduction
    Comparative Performance Evaluation of Cache-Coherent NUMA and COMA Architectures Per Stenstromt, Truman Joe, and Anoop Gupta Computer Systems Laboratory Stanford University, CA 94305 Abstract machine. Currently, many research groups are pursuing the design and construction of such multiprocessors [12, 1, 10]. Two interesting variations of large-scale shared-memory ma- As research has progressed in this area, two interesting vari- chines that have recently emerged are cache-coherent mm- ants have emerged, namely CC-NUMA (cache-coherent non- umform-memory-access machines (CC-NUMA) and cache- uniform memory access machines) and COMA (cache-only only memory architectures (COMA). They both have dis- memory architectures). Examples of the CC-NUMA ma- tributed main memory and use directory-based cache coher- chines are the Stanford DASH multiprocessor [12] and the MIT ence. Unlike CC-NUMA, however, COMA machines auto- Alewife machine [1], while examples of COMA machines are matically migrate and replicate data at the main-memoty level the Swedish Institute of Computer Science’s Data Diffusion in cache-line sized chunks. This paper compares the perfor- Machine (DDM) [10] and Kendall Square Research’s KSR1 mance of these two classes of machines. We first present a machine [4]. qualitative model that shows that the relative performance is Common to both CC-NUMA and COMA machines are the primarily determined by two factors: the relative magnitude features of distributed main memory, scalable interconnection of capacity misses versus coherence misses, and the gramr- network, and directory-based cache coherence. Distributed hirity of data partitions in the application. We then present main memory and scalable interconnection networks are es- quantitative results using simulation studies for eight prtraUeI sential in providing the required scalable memory bandwidth, applications (including all six applications from the SPLASH while directory-based schemes provide cache coherence with- benchmark suite).
    [Show full text]
  • Detailed Cache Coherence Characterization for Openmp Benchmarks
    1 Detailed Cache Coherence Characterization for OpenMP Benchmarks Anita Nagarajan, Jaydeep Marathe, Frank Mueller Dept. of Computer Science, North Carolina State University, Raleigh, NC 27695-7534 [email protected], phone: (919) 515-7889 Abstract models in their implementation (e.g., [6], [13], [26], [24], [2], [7]), and they operate at different abstraction levels Past work on studying cache coherence in shared- ranging from cycle-accuracy over instruction-level to the memory symmetric multiprocessors (SMPs) concentrates on operating system interface. On the performance tuning end, studying aggregate events, often from an architecture point work mostly concentrates on program analysis to derive op- of view. However, this approach provides insufficient infor- timized code (e.g., [15], [27]). Recent processor support for mation about the exact sources of inefficiencies in paral- performance counters opens new opportunities to study the lel applications. For SMPs in contemporary clusters, ap- effect of applications on architectures with the potential to plication performance is impacted by the pattern of shared complement them with per-reference statistics obtained by memory usage, and it becomes essential to understand co- simulation. herence behavior in terms of the application program con- In this paper, we concentrate on cache coherence simu- structs — such as data structures and source code lines. lation without cycle accuracy or even instruction-level sim- The technical contributions of this work are as follows. ulation. We constrain ourselves to an SPMD programming We introduce ccSIM, a cache-coherent memory simulator paradigm on dedicated SMPs. Specifically, we assume the fed by data traces obtained through on-the-fly dynamic absence of workload sharing, i.e., only one application runs binary rewriting of OpenMP benchmarks executing on a on a node, and we enforce a one-to-one mapping between Power3 SMP node.
    [Show full text]
  • Analysis and Optimization of I/O Cache Coherency Strategies for Soc-FPGA Device
    Analysis and Optimization of I/O Cache Coherency Strategies for SoC-FPGA Device Seung Won Min Sitao Huang Mohamed El-Hadedy Electrical and Computer Engineering Electrical and Computer Engineering Electrical and Computer Engineering University of Illinois University of Illinois California State Polytechnic University Urbana, USA Urbana, USA Ponoma, USA [email protected] [email protected] [email protected] Jinjun Xiong Deming Chen Wen-mei Hwu IBM T.J. Watson Research Center Electrical and Computer Engineering Electrical and Computer Engineering Yorktown Heights, USA University of Illinois University of Illinois [email protected] Urbana, USA Urbana, USA [email protected] [email protected] Abstract—Unlike traditional PCIe-based FPGA accelerators, However, choosing the right I/O cache coherence method heterogeneous SoC-FPGA devices provide tighter integrations for different applications is a challenging task because of between software running on CPUs and hardware accelerators. it’s versatility. By choosing different methods, they can intro- Modern heterogeneous SoC-FPGA platforms support multiple I/O cache coherence options between CPUs and FPGAs, but these duce different types of overheads. Depending on data access options can have inadvertent effects on the achieved bandwidths patterns, those overheads can be amplified or diminished. In depending on applications and data access patterns. To provide our experiments, we find using different I/O cache coherence the most efficient communications between CPUs and accelera- methods can vary overall application execution times at most tors, understanding the data transaction behaviors and selecting 3.39×. This versatility not only makes designers hard to decide the right I/O cache coherence method is essential.
    [Show full text]
  • Chapter 5 Thread-Level Parallelism
    Chapter 5 Thread-Level Parallelism Instructor: Josep Torrellas CS433 Copyright Josep Torrellas 1999, 2001, 2002, 2013 1 Progress Towards Multiprocessors + Rate of speed growth in uniprocessors saturated + Wide-issue processors are very complex + Wide-issue processors consume a lot of power + Steady progress in parallel software : the major obstacle to parallel processing 2 Flynn’s Classification of Parallel Architectures According to the parallelism in I and D stream • Single I stream , single D stream (SISD): uniprocessor • Single I stream , multiple D streams(SIMD) : same I executed by multiple processors using diff D – Each processor has its own data memory – There is a single control processor that sends the same I to all processors – These processors are usually special purpose 3 • Multiple I streams, single D stream (MISD) : no commercial machine • Multiple I streams, multiple D streams (MIMD) – Each processor fetches its own instructions and operates on its own data – Architecture of choice for general purpose mps – Flexible: can be used in single user mode or multiprogrammed – Use of the shelf µprocessors 4 MIMD Machines 1. Centralized shared memory architectures – Small #’s of processors (≈ up to 16-32) – Processors share a centralized memory – Usually connected in a bus – Also called UMA machines ( Uniform Memory Access) 2. Machines w/physically distributed memory – Support many processors – Memory distributed among processors – Scales the mem bandwidth if most of the accesses are to local mem 5 Figure 5.1 6 Figure 5.2 7 2. Machines w/physically distributed memory (cont) – Also reduces the memory latency – Of course interprocessor communication is more costly and complex – Often each node is a cluster (bus based multiprocessor) – 2 types, depending on method used for interprocessor communication: 1.
    [Show full text]
  • Reducing Cache Coherence Traffic with a NUMA-Aware Runtime Approach
    This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TPDS.2017.2787123, IEEE Transactions on Parallel and Distributed Systems IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. XX, NO. X, MONTH YEAR 1 Reducing Cache Coherence Traffic with a NUMA-Aware Runtime Approach Paul Caheny, Lluc Alvarez, Said Derradji, Mateo Valero, Fellow, IEEE, Miquel Moreto,´ Marc Casas Abstract—Cache Coherent NUMA (ccNUMA) architectures are a widespread paradigm due to the benefits they provide for scaling core count and memory capacity. Also, the flat memory address space they offer considerably improves programmability. However, ccNUMA architectures require sophisticated and expensive cache coherence protocols to enforce correctness during parallel executions, which trigger a significant amount of on- and off-chip traffic in the system. This paper analyses how coherence traffic may be best constrained in a large, real ccNUMA platform comprising 288 cores through the use of a joint hardware/software approach. For several benchmarks, we study coherence traffic in detail under the influence of an added hierarchical cache layer in the directory protocol combined with runtime managed NUMA-aware scheduling and data allocation techniques to make most efficient use of the added hardware. The effectiveness of this joint approach is demonstrated by speedups of 3.14x to 9.97x and coherence traffic reductions of up to 99% in comparison to NUMA-oblivious scheduling and data allocation. Index Terms—Cache Coherence, NUMA, Task-based programming models F 1 INTRODUCTION HE ccNUMA approach to memory system architecture the data is allocated within the NUMA regions of the system T has become a ubiquitous choice in the design-space [10], [28].
    [Show full text]
  • Parallel Processors and Cache Coherency
    Intro Cache Coherency Notes Computer Architecture II Parallel Processors and Cache Coherency Syed Asad Alam School of Computer Science and Statistics 1/31 Intro Cache Coherency Notes Introduction 2/31 Intro Cache Coherency Notes Flynn’s Taxonomy Introduced by M. J. Flynn SISD → Single processor with a single instruction stream and a single memory (Uniprocessors) SIMD → Single machine instruction with multiple processing elements operating on independent set of data MISD → Multiple processors, each of which execute different instruction sequence on a single sequence of data MIMD → Multiple processors execute different instruction sequences on different data sets 3/31 Intro Cache Coherency Notes Flynn’s Taxonomy Introduced by M. J. Flynn SISD → Single processor with a single instruction stream and a single memory (Uniprocessors) SIMD → Single machine instruction with multiple processing elements operating on independent set of data MISD → Multiple processors, each of which execute different instruction sequence on a single sequence of data MIMD → Multiple processors execute different instruction sequences on different data sets Text sourced from: Computer Organization and Architecture Designing for Performance, William Stallings, Chapter 17 & 18 3/31 Intro Cache Coherency Notes Flynn’s Taxonomy1 1 W. Stallings 4/31 Intro Cache Coherency Notes Flynn’s Taxonomy2 2 W. Stallings 5/31 Intro Cache Coherency Notes Multi-Computer/Cluster – Loosely Coupled 6/31 Intro Cache Coherency Notes Multi-Computer/Cluster – Loosely Coupled 6/31 Intro Cache Coherency Notes Cluster Computer Architecture3 3 W. Stallings 7/31 Intro Cache Coherency Notes Example IITAC Cluster [in Lloyd building] 346 x IBM e326 compute node each with 2 x 2.4GHz 64bit AMD Opteron 250 CPUs, 4GB RAM, ..
    [Show full text]
  • Comparative Modeling and Evaluation of CC-NUMA and COMA on Hierarchical Ring Rchitectures
    1316 IEEE TRANSACTIONSON PARALLEL AND DISTRIBUTEDSYSTEMS, VOL. 6, NO. 12, DECEMBER 1995 , Comparative Modeling and Evaluation of CC-NUMA and COMA on Hierarchical Ring rchitectures Xiaodong Zhang, Senior Member, IEEE, and Yong Yan Abstract-Parallel computing performance on scalable share& large NUMA architectures, the memory system organization memory architectures is affected by the structure of the intercon- also affects communication latency. With respect to the kinds nection networks linking processors to memory modules and on of memory organizations utilized, NUMA memory systems the efficiency of the memory/cache management systems. Cache Coherence Nonuniform Memory Access (CC-NUMA) and Cache can be classified into the following three types in terms of data Only Memory Access (COMA) are two effective memory systems, migration and coherence: and the hierarchical ring structure is an efficient interco~ection Non-CC-NUMA stands for. non-cache-coherentNUMA. network in hardware. This paper focuses on comparative per- formance modeling and evaluation of CC-NUMA and COMA on This type of architecture either supports no local caches a hierarchical ring shared-memory architecture. Analytical mod- at all (e.g., the BBN GPlOOO [I]), or provides a local. els for the two memory systems for comparative evaluation are cache that disallows caching of shared data in order to presented. Intensive performance measurements on data migra- avoid the cache coherence problem (e.g., the BBN tions have been conducted on the KSR-1, a COMA hlerarcbicaI TC2cKm[ 11). ring shared-memory machine. Experimental results support the analytical models, and we present practical observations and CCWUMA stands for cache-coherent NUMA, where comparisons of the two cache coherence memory systems.
    [Show full text]
  • 18-742 Parallel Computer Architecture Lecture 5: Cache Coherence
    18-742 Parallel Computer Architecture Lecture 5: Cache Coherence Chris Craik (TA) Carnegie Mellon University Readings: Coherence Required for Review Papamarcos and Patel, “A low-overhead coherence solution for multiprocessors with private cache memories,” ISCA 1984. Kelm et al, “Cohesion: A Hybrid Memory Model for Accelerators”, ISCA 2010. Required Censier and Feautrier, “A new solution to coherence problems in multicache systems,” IEEE Trans. Comput., 1978. Goodman, “Using cache memory to reduce processor-memory traffic,” ISCA 1983. Laudon and Lenoski, “The SGI Origin: a ccNUMA highly scalable server,” ISCA 1997. Lenoski et al, “The Stanford DASH Multiprocessor,” IEEE Computer, 25(3):63-79, 1992. Martin et al, “Token coherence: decoupling performance and correctness,” ISCA 2003. Recommended Baer and Wang, “On the inclusion properties for multi-level cache hierarchies,” ISCA 1988. Lamport, “How to Make a Multiprocessor Computer that Correctly Executes Multiprocess Programs”, IEEE Trans. Comput., Sept 1979, pp 690-691. Culler and Singh, Parallel Computer Architecture, Chapters 5 and 8. 2 Shared Memory Model Many parallel programs communicate through shared memory Proc 0 writes to an address, followed by Proc 1 reading This implies communication between the two Proc 0 Proc 1 Mem[A] = 1 … Print Mem[A] Each read should receive the value last written by anyone This requires synchronization (what does last written mean?) What if Mem[A] is cached (at either end)? 3 Cache Coherence Basic question: If multiple processors cache
    [Show full text]
  • Why On-Chip Cache Coherence Is Here to Stay
    University of Pennsylvania ScholarlyCommons Departmental Papers (CIS) Department of Computer & Information Science 7-2012 Why On-Chip Cache Coherence is Here to Stay Milo Martin University of Pennsylvania, [email protected] Mark D. Hill University of Wisconsin Daniel J. Sorin Duke University Follow this and additional works at: https://repository.upenn.edu/cis_papers Part of the Computer Sciences Commons Recommended Citation Milo Martin, Mark D. Hill, and Daniel J. Sorin, "Why On-Chip Cache Coherence is Here to Stay", . July 2012. Martin, M., Hill, M., & Sorin, D., Why On-Chip Cache Coherence is Here to Stay, Communications of the ACM, July 2012, doi: 10.1145/2209249.2209269 © 1994, 1995, 1998, 2002, 2009 by ACM, Inc. Permission to copy and distribute this document is hereby granted provided that this notice is retained on all copies, that copies are not altered, and that ACM is credited when the material is used to form other copyright policies. This paper is posted at ScholarlyCommons. https://repository.upenn.edu/cis_papers/708 For more information, please contact [email protected]. Why On-Chip Cache Coherence is Here to Stay Abstract Today’s multicore chips commonly implement shared memory with cache coherence as low-level support for operating systems and application software. Technology trends continue to enable the scaling of the number of (processor) cores per chip. Because conventional wisdom says that the coherence does not scale well to many cores, some prognosticators predict the end of coherence. This paper refutes this conventional wisdom by showing one way to scale on-chip cache coherence with bounded costs by combining known techniques such as: shared caches augmented to track cached copies, explicit cache eviction notifications, and hierarchical design.
    [Show full text]
  • Evaluating Cache Coherent Shared Virtual Memory for Heterogeneous Multicore Chips
    Evaluating Cache Coherent Shared Virtual Memory for Heterogeneous Multicore Chips BLAKE A. HECHTMAN , Duke University DANIEL J. SORIN , Duke University The trend in industry is towards heterogeneous multicore processors (HMCs), including chips with CPUs and massively-threaded throughput-oriented processors (MTTOPs) such as GPUs. Although current homogeneous chips tightly couple the cores with cache-coherent shared virtual memory (CCSVM), this is not the communication paradigm used by any current HMC. In this paper, we present a CCSVM design for a CPU/MTTOP chip, as well as an extension of the pthreads programming model, called xthreads, for programming this HMC. Our goal is to evaluate the potential performance benefits of tightly coupling heterogeneous cores with CCSVM. 1 INTRODUCTION The trend in general-purpose chips is for them to consist of multiple cores of various types— including traditional, general-purpose compute cores (CPU cores), graphics cores (GPU cores), digital signal processing cores (DSPs), cryptography engines, etc.—connected to each other and to a memory system. Already, general-purpose chips from major manufacturers include CPU and GPU cores, including Intel’s Sandy Bridge [42][16], AMD’s Fusion [4], and Nvidia Research’s Echelon [18]. IBM’s PowerEN chip [3] includes CPU cores and four special-purpose cores, including accelerators for cryptographic processing and XML processing. In Section 2, we compare current heterogeneous multicores to current homogeneous multicores, and we focus on how the cores communicate with each other. Perhaps surprisingly, the communication paradigms in emerging heterogeneous multicores (HMCs) differ from the established, dominant communication paradigm for homogeneous multicores. The vast majority of homogeneous multicores provide tight coupling between cores, with all cores communicating and synchronizing via cache-coherent shared virtual memory (CCSVM).
    [Show full text]
  • A CMOS Vector Processor with a Custom Streaming Cache Greg Faanes [email protected] Silicon Graphics
    A CMOS Vector Processor with a Custom Streaming Cache Greg Faanes [email protected] Silicon Graphics A CMOS Vector Processor with a Custom Streaming Cache G. Faanes Hot Chips 98 1 A CMOS Vector Processor With a Custom Streaming Cache Overview: • Goals and ground rules for the design • CPU/Cache overview • CPU chip • Scalar unit design • Vector unit design • Streaming cache chip • Cache unit design • Summary A CMOS Vector Processor with a Custom Streaming Cache G. Faanes Hot Chips 98 2 Goals and Ground Rules for the Design • Built to run large scale scientific applications • Good vector performance • High bandwidth to cache and main memory • unit stride • non-unit stride • gather/scatter • YMP upward compatible • YMP ⇒ J90 ⇒ J90se ⇒ SV1 processor • Few resources • 3 logic designers • “Off the shelf” technology A CMOS Vector Processor with a Custom Streaming Cache G. Faanes Hot Chips 98 3 Not Your Ordinary Processor • Not built to run SPECINT95 fast • Different cache and system interface • 1 set of pins for both cache and memory • High performance through multiple parallel and pipelined functional units A CMOS Vector Processor with a Custom Streaming Cache G. Faanes Hot Chips 98 4 SV1 Chipset CPU Chip Cache Chip Clock Frequency 250-300 Mhz 250-300 Mhz Performance 1.0-1.2 GFLOP/sec 128 KB and 4.0-4.8 GB/sec Die Size 14.6 mm x 14.6 mm = 213 mm2 13.7 mm x 13.7 mm = 188 mm2 Technology CMOS, 5 layer metal, 2.6 v CMOS, 5 layer metal, 2.6 v Transistors 9.7 million 14.9 million Power 12 watts 8 watts Signal Pins 635 601 A CMOS Vector Processor with a Custom Streaming Cache G.
    [Show full text]
  • Computer Architecture, Part 7
    Part VII Advanced Architectures Feb. 2011 Computer Architecture, Advanced Architectures Slide 1 About This Presentation This presentation is intended to support the use of the textbook Computer Architecture: From Microprocessors to Supercomputers, Oxford University Press, 2005, ISBN 0-19-515455-X. It is updated regularly by the author as part of his teaching of the upper-division course ECE 154, Introduction to Computer Architecture, at the University of California, Santa Barbara. Instructors can use these slides freely in classroom teaching and for other educational purposes. Any other use is strictly prohibited. © Behrooz Parhami Edition Released Revised Revised Revised Revised First July 2003 July 2004 July 2005 Mar. 2007 Feb. 2011* * Minimal update, due to this part not being used for lectures in ECE 154 at UCSB Feb. 2011 Computer Architecture, Advanced Architectures Slide 2 VII Advanced Architectures Performance enhancement beyond what we have seen: • What else can we do at the instruction execution level? • Data parallelism: vector and array processing • Control parallelism: parallel and distributed processing Topics in This Part Chapter 25 Road to Higher Performance Chapter 26 Vector and Array Processing Chapter 27 Shared-Memory Multiprocessing Chapter 28 Distributed Multicomputing Feb. 2011 Computer Architecture, Advanced Architectures Slide 3 25 Road to Higher Performance Review past, current, and future architectural trends: • General-purpose and special-purpose acceleration • Introduction to data and control parallelism Topics in This Chapter 25.1 Past and Current Performance Trends 25.2 Performance-Driven ISA Extensions 25.3 Instruction-Level Parallelism 25.4 Speculation and Value Prediction 25.5 Special-Purpose Hardware Accelerators 25.6 Vector, Array, and Parallel Processing Feb.
    [Show full text]