Cisco Technology, Inc.

Allstate Solutions Private Limited

Amit Kumar Saha

Debojyoti Dutta

Johnu George

Komei Shimamura

Xinyuan Huang

Ralf Rantzau

Arun Saha

Madhu S Kumar

Ramdoot Kumar Pydipaty

Sourav Chakraborty

Marc Solanas Tarre

Timothy Okwii

Adhita Selvaraj

Aspects of the subject technology relate to ways to determine the optimal storage of data structures in a hierarchy of memory types. In some aspects, a process of the technology can include steps for identifying a retrieval cost associated with retrieving a field in an object from data storage, comparing the retrieval cost for the field to a cost threshold for storing data in persistent memory, and selectively storing the field in either a persistent memory device or a non-persistent memory device based on a comparison of the retrieval cost for the field to the cost threshold. Systems and machine-readable media are also provided.

Intelligent layout of composite data structures in tiered storage with persistent memory

Systems, methods, and computer-readable media for managing storing of data in a data storage system using a client tag. In some examples, a first portion of a data load as part of a transaction and a client identifier that uniquely identifies a client is received from the client at a data storage system. The transaction can be tagged with a client tag including the client identifier and the first portion of the data load can be stored in storage at the data storage system. A first log entry including the client tag is added to a data storage log in response to storing the first portion of the data load in the storage. The first log entry is then written from the data storage log to a persistent storage log in persistent memory which is used to track progress of storing the data load in the storage.

Using persistent memory to enable restartability of bulk load transactions in cloud databases

A method can include receiving, at a workflow controller, a machine learning workflow, the machine learning workflow associated with a first task and a second task. The first task is training a machine learning model and the second task is deploying the model. The method can include segmenting, by the workflow controller, the machine learning workflow into a first sub-workflow associated with the first task and a second sub-workflow associated with the second task, assigning a first workflow agent to the first sub-workflow and assigning a second workflow agent to the second sub-workflow, selecting, by the first workflow agent and based on first resources needed to perform the first task, a first cluster for performing the first task and selecting, by the second workflow agent and based on second resources needed to perform the second task, a second cluster for performing the second task.

Systems and methods for providing management of machine learning components

Methods and systems for automatically generating validation scripts for software applications are disclosed. A computing device may receive a first application. The computing device may compare the first application to a plurality of stored applications. The computing device may determine a second application among the plurality of stored applications based on the comparing. The computing device may determine a first validation script associated with the second application. The computing device may automatically generate a second validation script for the first application based on the first validation script and a result of a comparison of the first application and the second application. The computing device may validate the first application using the second validation script.

Automated validation script generation and execution engine

In one embodiment, a method for FPGA accelerated serverless computing comprises receiving, from a user, a definition of a serverless computing task comprising one or more functions to be executed. A task scheduler performs an initial placement of the serverless computing task to a first host determined to be a first optimal host for executing the serverless computing task. The task scheduler determines a supplemental placement of a first function to a second host determined to be a second optimal host for accelerating execution of the first function, wherein the first function is not able to accelerated by one or more FPGAs in the first host. The serverless computing task is executed on the first host and the second host according to the initial placement and the supplemental placement.

FPGA acceleration for serverless computing

A method for accelerating data operations across a plurality of nodes of one or more clusters of a distributed computing environment. Rack awareness information characterizing the plurality of nodes is retrieved and a non-volatile memory (NVM) capability of each node is determined. A write operation is received at a management node of the plurality of nodes and one or more of the rack awareness information and the NVM capability of the plurality of nodes are analyzed to select one or more nodes to receive at least a portion of the write operation, wherein at least one of the selected nodes has an NVM capability. A multicast group for the write operation is then generated wherein the selected nodes are subscribers of the multicast group, and the multicast group is used to perform hardware accelerated read or write operations at one or more of the selected nodes.

Accelerating data replication using multicast and non-volatile memory enabled nodes

A method for data provisioning a serverless computing cluster. A plurality of user defined functions (UDFs) are received for execution on worker nodes of the serverless computing cluster. For a first UDF, one or more data locations of UDF data needed to execute the first UDF are determined. At a master node of the serverless computing cluster, a plurality of worker node tickets are received, each ticket indicating a resource availability of a corresponding worker node. The one or more data locations and the plurality of worker node tickets are analyzed to determine eligible worker nodes capable of executing the first UDF. The master node transmits a pre-fetch command to one or more of the eligible worker nodes, causing the eligible worker nodes to become a provisioned worker node for the first UDF by storing a pre-fetched first UDF data before the first UDF is assigned for execution.

Provisioning using pre-fetched data in serverless computing environments

One aspect of the disclosure relates to, among other things, a method for optimizing and provisioning a software-as-a-service (SaaS). The method includes determining a graph comprising interconnected stages for the SaaS, wherein each stage has a replication factor and one or more metrics that are associated with one or more service level objectives of the SaaS, determining a first replication factor associated with a first one of the stages which meets a first service level objective of the SaaS, adjusting the first replication factor associated with the first one of the stage based on the determined first replication factor, and provisioning the SaaS onto networked computing resources based on the graph and replication factors associated with each stage.

Estimating model parameters for automatic deployment of scalable micro services

Systems, methods, and computer-readable storage media are provided for storing machine learned models in a tiered storage. The model serving network evaluates where the models should be stored based on the model corresponding service level agreement. The model is generally stored at the lowest tiered storage device that is still capable of satisfying the model's service level agreement. In this way, the model serving network aims to store data that achieves the cheapest cost.

Using tiered storage and ISTIO to satisfy SLA in model serving and updates

Systems, methods, and computer-readable media for replicating data in a distributed storage cluster using an underlying network. In some examples, a primary node of a placement group in a network overlay of a distributed storage cluster can receive data for replication in the placement group. The primary node can provide the data to a first slave node of a plurality of slave nodes within the placement group in an underlying network of the distributed storage cluster. The data can subsequently be replicated using the underlying network by providing the data to at least one other slave node of the plurality of slave nodes within the placement group in the underlying network directly from the first slave node in the underlying network.

Using network device replication in distributed storage clusters

Aspects of the disclosed technology relate to ways to determine the optimal storage of data structures across different memory device is associated with physically disparate network nodes. In some aspects, a process of the technology can include steps for receiving a first retrieval request for a first object, searching a local PMEM device for the first object based on the first retrieval request, in response to a failure to find the first object on the local PMEM device, transmitting a second retrieval request to a remote node, wherein the second retrieval request is configured to cause the remote node to retrieve the first object from a remote PMEM device. Systems and machine-readable media are also provided.

Accessing composite data structures in tiered storage across network nodes

In one embodiment, a method implements virtualized network functions in a serverless computing system having networked hardware resources. An interface of the serverless computing system receives a specification for a network service including a virtualized network function (VNF) forwarding graph (FG). A mapper of the serverless computing system determines an implementation graph comprising edges and vertices based on the specification. A provisioner of the serverless computing system provisions a queue in the serverless computing system for each edge. The provisioner further provisions a function in the serverless computing system for each vertex, wherein, for at least one or more functions, each one of said at least one or more functions reads incoming messages from at least one queue. The serverless computing system processes data packets by the queues and functions in accordance with the VNF FG. The queues and functions processes data packets in accordance with the VNF FG.

Growing community of inventors

Bengaluru, India

Amit Kumar Saha