… In this paper, we propose Finesse, a fast resemblancedetection approach that exploits the …
each chunk into several subchunks and then quickly computing features from each subchunk, …

Graph processing is becoming commonplace in many applications to analyze huge datasets.
Much of the prior work in this area has assumed I/O devices with considerable latencies, …

… to use fast NVM devices at their full potential. As an embedded store, uDepot’s performance
nearly matches the raw performance of fast … These devices, which we refer to as Fast NVMe …

… Proceedings of the 17th USENIX Conference on File and Storage Technologies (FAST ’19). …
The author informed us that the two papers used different chips of the same manufacturer. …

… We also compare Orion to ext4 mounted on a remote iSCSI target hosting a ramdisk using
iSCSI Extension over RDMA (iSER) [12] (denoted by Ext4/iSER), which provides the client …

… Let us now see how we traverse the directories. Note that the local depth of each segment
and the global depth determine the segment’s stride in the directory, ie, how many times the …

… , targeting disk is faster, since Ziggurat can buffer the data in DRAM more quickly than it can
… Hot files and hot data in unevenly-accessed files remain in the faster tier. When the usage of …

… This paper presents MITOSIS, an operating system primitive that provides fast remote fork,
which exploits a deep codesign of the OS kernel with RDMA. By leveraging the fast remote …

… For the Backblaze dataset of 100,000+ HDDs over 5 years, our analysis shows that using
HeART’s settings could achieve data reliability targets with 11–33% fewer HDDs, depending …

… Here we demonstrate the problems associated with this approach, using the world’s No.3
supercomputer TaihuLight as a sample platform. Specifically, we discuss resource …