Qing  Zheng Presents at 38th International Conference on Massive Storage Systems and Technology

Qing  Zheng Presents at 38th International Conference on Massive Storage Systems and Technology

Qing  Zheng Presents at 38th International Conference on Massive Storage Systems and Technology

New Mexico Consortium scientist Qing Zheng recently presented his work titled, Analytics Query Pushdown Using Object-Based Computational Storage at the 38th International Conference on Massive Storage Systems and Technology.

This conference, held Jun 3-7th in Santa Clara, CA, was founded, in 1974, by the leading national laboratories, MSST (the International Conference on Massive Storage Systems and Technology) has been a venue for massive-scale storage system designers and implementers, storage architects, researchers, and vendors to share best practices and discuss building and securing the world’s largest storage systems for high-performance computing, web-scale systems, and enterprises.

Open standards facilitate interoperability, community support, and vendor neutrality, ensuring that diverse systems and applications can work together seamlessly, fostering innovation through collaborative development, and preventing vendor lock-in by providing multiple sources of compatible products and solutions. Just as NFS sets the protocol for NAS and ANSI T10 defines SCSI’s Object-based Storage Device (OSD) command set, this talk pushes for standardizing object-based computational storage amid rising object storage use, the increasing bottleneck caused by excessive data movement, a lack of standardized methods for integrating data reduction functions within storage, and a shift towards NVMe as a modern replacement for the old SCSI interface.

As a collaborative effort between SK hynix, Los Alamos National Laboratory, Versity, Neuroblade, and Airmettle, Qing and his colleagues envision an open, standard Object-based Computational Storage (OCS) stack as a new computational storage platform for data analytics.

This stack comprises a high-level Object-based Computational Storage (OCS) interface and a low-level Object-based Computational Storage Device (OCSD) command set for NVMe-based storage devices or arrays. A typical setup would consist of a pool of gateway servers implementing the high-level interface and NVMe devices or arrays implementing the low-level interface.

Qing’s presentation focused on the rationale behind such an open storage stack, its integration with existing storage ecosystems, and its current prototype implementation along with early analytics acceleration results using real-world workload and dataset.

To learn more about this conference see: https://www.msstconference.org/