An official website of the United States government
After over a decade of researcher anticipation for the arrival of persistent memory (PMem), the first shipments of 3D XPoint-based Intel Optane Memory in 2019 were quickly followed by its cancellation in 2022. Was this another case of an idea quickly fading from future to past tense, relegating work in this area to the graveyard of failed technologies? The recently introduced Compute Express Link (CXL) may offer a path forward, with its persistent memory profile offering a universal PMem attachment point. Yet new technologies for memory-speed persistence seem years off, and may never become competitive with evolving DRAM and flash speeds. Without persistent memory itself, is future PMem research doomed? We offer two arguments for why reports of the death of PMem research are greatly exaggerated. First, the bulk of persistent-memory research has not in fact addressed memory persistence, but rather in-memory crash consistency, which was never an issue in prior systems where CPUs could not observe post-crash memory states. CXL memory pooling allows multiple hosts to share a single memory, all in different failure domains, raising crash-consistency issues even with volatile memory. Second, we believe CXL necessitates a ``disaggregation'' of PMem research. Most work to date assumed a single technology and set of features, \ie speed, byte addressability, and CPU load/store access. With an open interface allowing new topologies and diverse PMem technologies, we argue for the need to examine these features individually and in combination. While one form of PMem may have been canceled, we argue that the research problems it raised not only remain relevant but have expanded in a CXL-based future.
more »
« less
Zhuque: Failure is Not an Option, it's an Exception
Persistent memory (PMEM) allows direct access to fast storage at byte granularity. Previously, processor caches backed by persistent memory were not persistent, complicating the design of persistent applications and reducing their performance. A new generation of systems with flush-on-fail semantics effectively offer persistent caches, offering the potential for much simpler, faster PMEM programming models. This work proposes Whole Process Persistence (WPP), a new programming model for systems with persistent caches. In the WPP model, all process state is made persistent. On restart after power failure, this state is reloaded and execution resumes in an application-defined interrupt handler. We also describe the Zhuque runtime, which transparently provides WPP by interposing on the C bindings for system calls in userspace. It requires little or no programmer effort to run applications on Zhuque. Our measurements show that Zhuque outperforms state of the art PMEM libraries, demonstrating mean speedups across all benchmarks of 5.24x over PMDK, 3.01x over Mnemosyne, 5.43x over Atlas, and 4.11x over Clobber-NVM. More important, unlike existing systems, Zhuque places no restrictions on how applications implement concurrency, allowing us to run a newer version of Memcached on Zhuque and gain more than 7.5x throughput over the fastest existing persistent implementations.
more »
« less
- Award ID(s):
- 2011213
- PAR ID:
- 10450322
- Editor(s):
- Lawall, Julia; Williams, Dan
- Date Published:
- Journal Name:
- Proceedings of the 2023 USENIX Annual Technical Conference
- Issue:
- 2023
- Page Range / eLocation ID:
- 833-849
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)High capacity persistent memory (PMEM) is finally commercially available in the form of Intel's Optane DC Persistent Memory Module (DCPMM). Researchers have raced to evaluate and understand the performance of DCPMM itself as well as systems and applications designed to leverage PMEM resulting from over a decade of research. Early evaluations of DCPMM show that its behavior is more nuanced and idiosyncratic than previously thought. Several assumptions made about its performance that guided the design of PMEM-enabled systems have been shown to be incorrect. Unfortunately, several peculiar performance characteristics of DCPMM are related to the memory technology (3D-XPoint) used and its internal architecture. It is expected that other technologies (such as STT-RAM, memristor, ReRAM, NVDIMM), with highly variable characteristics, will be commercially shipped as PMEM in the near future. Current evaluation studies fail to understand and categorize the idiosyncratic behavior of PMEM; i.e., how do the peculiarities of DCPMM related to other classes of PMEM. Clearly, there is a need for a study which can guide the design of systems and is agnostic to PMEM technology and internal architecture. In this paper, we first list and categorize the idiosyncratic behavior of PMEM by performing targeted experiments with our proposed PMIdioBench benchmark suite on a real DCPMM platform. Next, we conduct detailed studies to guide the design of storage systems, considering generic PMEM characteristics. The first study guides data placement on NUMA systems with PMEM while the second study guides the design of lock-free data structures, for both eADR- and ADR-enabled PMEM systems. Our results are often counter-intuitive and highlight the challenges of system design with PMEM.more » « less
-
Hardware Transactional Memory (HTM) simplifies concurrent programming and can accelerate multithreaded execution through lock elision. Non-Volatile Memory (NVM) combines the speed and byte addressability of DRAM with the durability of storage, enabling the construction of high-performance, persistent data structures. Unfortunately, the write-back instructions typically needed to ensure post-crash consistency in NVM cause HTM transactions to abort, precluding the straightforward combination of HTM and persistent data structures. The problem goes away on machines with persistent caches, but these require special battery-backed circuitry and are far from commonplace.To combine HTM and persistent data structures, we advocate for buffered durable linearizability (BDL), a relaxed correctness criterion that enables recovery to a "recent" consistent state in the wake of a crash, allowing writes-back to occur outside transactions.Significantly, BDL retains the persistence guarantees of storage systems—such as databases backed by disks or flash—that have relied on buffering for decades.The combination of HTM and buffered durability enables three separate usage scenarios. First, we add durability to an existing HTM-based structure (a van Emde Boas tree due to Khalaji et al.); second, we use HTM to simplify an existing persistent structure (a skiplist due to Wang et al.); third, we "back port" an HTM-based structure optimized for persistent caches (a hash table due to Zhang et al.) to work well on more conventional processors. The first two scenarios yield several-fold improvements in throughput; the third sees very little slowdown.more » « less
-
The memory wall challenge -- the growing disparity between CPU speed and memory speed -- has been one of the most critical and long-standing challenges in computing. For high performance computing, programming to achieve efficient execution of parallel applications often requires more tuning and optimization efforts to improve data and memory access than for managing parallelism. The situation is further complicated by the recent expansion of the memory hierarchy, which is becoming deeper and more diversified with the adoption of new memory technologies and architectures such as 3D-stacked memory, non-volatile random-access memory (NVRAM), and hybrid software and hardware caches. The authors believe it is important to elevate the notion of memory-centric programming, with relevance to the compute-centric or data-centric programming paradigms, to utilize the unprecedented and ever-elevating modern memory systems. Memory-centric programming refers to the notion and techniques of exposing hardware memory system and its hierarchy, which could include DRAM and NUMA regions, shared and private caches, scratch pad, 3-D stacked memory, non-volatile memory, and remote memory, to the programmer via portable programming abstractions and APIs. These interfaces seek to improve the dialogue between programmers and system software, and to enable compiler optimizations, runtime adaptation, and hardware reconfiguration with regard to data movement, beyond what can be achieved using existing parallel programming APIs. In this paper, we provide an overview of memory-centric programming concepts and principles for high performance computing.more » « less
-
We consider a parallel computational model, the Parallel Persistent Memory model, comprised of P processors, each with a fast local ephemeral memory of limited size, and sharing a large persistent memory. The model allows for each processor to fault at any time (with bounded probability), and possibly restart. When a processor faults, all of its state and local ephemeral memory is lost, but the persistent memory remains. This model is motivated by upcoming non-volatile memories that are nearly as fast as existing random access memory, are accessible at the granularity of cache lines, and have the capability of surviving power outages. It is further motivated by the observation that in large parallel systems, failure of processors and their caches is not unusual. We present several results for the model, using an approach that breaks a computation into capsules, each of which can be safely run multiple times. For the single-processor version we describe how to simulate any program in the RAM, the external memory model, or the ideal cache model with an expected constant factor overhead. For the multiprocessor version we describe how to efficiently implement a work-stealing scheduler within the model such that it handles both soft faults, with a processor restarting, and hard faults, with a processor permanently failing. For any multithreaded fork-join computation that is race free, write-after-read conflict free and has W work, D depth, and C maximum capsule work in the absence of faults, the scheduler guarantees a time bound on the model of O(W/P_A+ (DP/P_A ) log_{1/(Cf )} W) in expectation, where P is the maximum number of processors, P_A is the average number, and f ≤ 1/(2C) is the probability a processor faults between successive persistent memory accesses. Within the model, and using the proposed methods, we develop efficient algorithms for parallel prefix sums, merging, sorting, and matrix multiply.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
