Twizzler: A
            <i>Data-centric</i>
            OS for Non-volatile Memory

Bittman, Daniel; Alvaro, Peter; Mehra, Pankaj; Long, Darrell D. E.; Miller, Ethan L.

doi:10.1145/3454129

Cited by 7 publications

(5 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pointer swizzling is one of the most related techniques. Prior studies [18,43] swizzle persistent references to virtual addresses. However, they do not tackle the validity issue on data movement.…”

Section: Related Workmentioning

confidence: 99%

PMLiteDB: Streamlining Access Paths for High-Performance Persistent Memory Document Database Systems

Jin

Wei

Sha

et al. 2023

IEEE Trans. Comput.

View full text Add to dashboard Cite

This paper points out an important challenge for document databases to enjoy the fast byte-addressable persistent memory, such that excessive data indexing and movement become the performance bottlenecks. The advent of byte-addressable persistent memory opens an important opportunity for document databases to read and write durable data fetching them into DRAM. Reaping the benefit of persistent memory is not straightforward, as existing document databases are tailored for disk storage. They assume that the disk and DRAM data movement dominates the performance. However, data indexing becomes the performance bottleneck when porting the document databases to persistent memory. The paper proposes PMLiteDB, the first persistent memory document database with streamlined access paths. PMLiteDB introduces two techniques, direct reading and selective caching. Direct reading streamlines the translation from document IDs to the address of documents whenever possible by swizzling the IDs into persistent memory(PM) references. It guarantees to use only up-to-date PM references when document movements invalidate associated references. Selective caching reduces data movements between DRAM and persistent memory by selectively caching persistent memory data page with DRAM buffer. For accesses on other pages, the mechanism directly load the data on persistent memory. Compared to the design that adopts persistent memory as a fast disk without exploiting the byte-addressability, PMLiteDB achieves 2.33× on average and up to 6.18× speedup.

show abstract

Section: Related Workmentioning

confidence: 99%

PMLiteDB: Streamlining Access Paths for High-Performance Persistent Memory Document Database Systems

Jin

Wei

Sha

et al. 2023

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…Even though the Intel Optane DC Persistent DIMM can provide significant benefits for future systems due to its performance characteristics, there are several challenges to solve before leveraging such devices in future systems, including: (1) the need for system mechanisms for proper data placement between DRAM and the Optane DIMM device [93], [94], [95], [96], [114], [191], (2) difficulties in fabricating printed circuit boards (PCBs) for mobile platforms that can accommodate the Optane DIMM, and (3) accommodating the high cost-per-bit of the Intel Optane DIMM in cost-sensitive mobile systems. Even though prior works [131], [132], [133], [134], [136], [137] propose, using simulation models, to integrate byte-addressable NVM devices as swap space for mobile systems, we choose the Intel Optane SSD for our studies for two main reasons.…”

Section: Intel Optane Ssdmentioning

confidence: 99%

Extending Memory Capacity in Modern Consumer Systems With Emerging Non-Volatile Memory: Experimental Analysis and Characterization Using the Intel Optane SSD

Oliveira,

Ghose,

Gómez-Luna

et al. 2023

IEEE Access

View full text Add to dashboard Cite

DRAM scalability is becoming a limiting factor to the available memory capacity in consumer devices. As a potential solution, manufacturers have introduced emerging non-volatile memories (NVMs) into the market, which can be used to increase the memory capacity of consumer devices by augmenting or replacing DRAM. In this work, we provide the first analysis of the impact of extending the main memory space of consumer devices using off-the-shelf NVMs. We equip real web-based Chromebook computers with the Intel Optane solid-state drive (SSD), which contains state-of-the-art low-latency NVM, and use the NVM as swap space. We analyze the performance and energy consumption of the Optane-equipped Chromebooks, and compare this with (i) a baseline system with double the amount of DRAM than the system with the NVM-based swap space; and (ii) a system where the Intel Optane SSD is naively replaced with a state-of-theart NAND-flash-based SSD. Our experimental analysis reveals that while Optane-based swap space provides a cost-effective way to alleviate the DRAM capacity bottleneck in consumer devices, naive integration of the Optane SSD leads to several system-level overheads, mostly related to (1) the Linux block I/O layer, which can negatively impact overall performance; and (2) the off-chip traffic to the swap space, which can negatively impact energy consumption. To reduce the Linux block I/O layer overheads, we tailor several system-level mechanisms (i.e., the I/O scheduler and the I/O request completion mechanism) to the currentlyrunning application's access pattern. To reduce the off-chip traffic overhead, we leverage an operating system feature (called Zswap) that allocates some DRAM space to be used as a compressed in-DRAM cache for data swapped between DRAM and the Intel Optane SSD, significantly reducing energy consumption caused by the off-chip traffic to the swap space. We conclude that emerging NVMs are a cost-effective solution to alleviate the DRAM capacity bottleneck in consumer devices, which can be further enhanced by tailoring system-level mechanisms to better leverage the characteristics of our workloads and the NVM.

show abstract

“…New device-side software is drawing upon the analogy between memory and storage and building for disaggregated memory what services such as S3 built for cloud storage: a foundation based on self-contained objects. 7 In these new products, memory objects rescue the translation context required by graph-structured data and compute and embed the necessary information in the form of a foreign object table that resides at a known location in every memory object.…”

Section: Industry's Road Map Of Memory Disaggregationmentioning

confidence: 99%