Florida International University

Raju Rangaswami

Chi Zhang

Evangelos Christidis

S Masoud Sadjadi

Yi Deng

Steve Luis

Peter Clarke

Adnan Maruf

Janki Bhimani

Ricardo Koller

Ashikee Ghosh

Saman Biook Aghazadeh

Daniel Campello

Steven Lyons

Luis Useche

Hector Lopez

Systems and methods for dynamically placing data in a hybrid memory structure are provided. A machine learning (ML)-based, adaptive tiered memory system can actively monitor application memory to dynamically place the right data in the right memory tier at the right time. The memory system can use reinforcement learning to perform dynamic tier placement of memory pages.

Machine learning based tiered memory systems and methods

Techniques and systems are disclosed for implementing non-blocking writes to eliminate the fetch-before-write requirement by creating an in-memory patch for the updated page and unblocking the calling process. Non-blocking writes eliminate such blocking by buffering the written data elsewhere in memory and unblocking the writing process immediately. Subsequent reads to the updated page locations are also made non-blocking and, in some cases, can be eliminated when the read request can be serviced from in-memory patches. Implementation scenarios can include an operating system (OS) enhancement, revision to an existing OS component (e.g., the OS kernel), special OS component, or enhancement to the software or firmware of the controller software or microcontroller of a storage device or array of storage devices.

Non-blocking writes to file data

Techniques and systems are described for enabling local independent failure domains in a host server or datacenter environment. Included is a locally-attached independent failure device (LA-IFD) with an independent data buffer and a local communications bus for attaching to a host server. Techniques for a communications protocol between the LA-IFD and its host server are provided, including: receiving a request to write a data segment to persistent storage; synchronously storing the data segment in a buffered data segment at the LA-IFD and initiating an asynchronous update of the data segment at a remote storage system; sending a write acknowledgement indicating completion to the requestor; and, after receiving a completion notification from the remote storage system, removing the buffered data segment from the LA-IFD. In some cases, techniques allow a host server and LA-IFD pair to monitor one another for failures and implement a modified protocol in the event of unavailability.

Techniques and systems for local independent failure domains

A system for automating at least a portion of a communication application development cycle using a communication virtual machine that automatically interfaces with an existing network system and provides an engine for executing designed application models over the existing network system.

Communication virtual machine

Growing community of inventors

Miami, FL, United States of America

Raju Rangaswami