352071BR
Your Role and Responsibilities
Do you want to work with top cloud researcher? We offer a variety of summer internships in the cloud area.
Address space separation inside Linux Kernel
Since the invention of the virtual memory, programs are protected from each other through address space isolation. As kernel exploits and speculation vulnerabilities are inevitable, it's worth to add kernel space isolation to minimize damage that these exploits can cause. IBM leads Linux kernel community discussion to explore and realize these ideas. Initial set of patches have been posted to mailing lists and presented in several developer conferences in 2019. As part of this internship, a student will have an opportunity to take part in this forward-looking exploration, validating the existing and generating new approaches for full stack secure isolation between Linux workloads, e.g. VMs, containers, or processes.
AIOps for Cloud and OpenShift Infrastructure
The Hybrid Cloud Infrastructure team at IBM Research Haifa works at the cutting edge of applied research in distributed systems and AI. We bring distributed systems and data science expertise to IBM’s cloud Infrastructure assets and develop novel approaches to tackle real-world problems.
In this project you will work on applying AI to the big data of operations (logs, metrics, tickets, etc.) to automate the operations of the cloud infrastructure. This is an opportunity to develop AI techniques using the huge amount of operations data coming from IBM’s cloud and OpenShift. A candidate topic is improving our multivariate anomaly detector and root cause analysis tool by exploring additional/alternative deep learning and machine learning techniques.
Building the storage stack of tomorrow: advanced features for Ceph
Ceph is an open source distributed storage system that provides a scalable unified storage for block, file and object storage. Ceph is widely used with Kubernetes and OpenShift deployments.
The internship is focused on developing advanced storage functions for the Ceph block interface.
The block interface, called RBD, is the most widely used Ceph interface and offers a reliable, highly scalable and perfomant storage system. The aim is to enhance the features of the Ceph platform related to security, reliability and performance. For example, in the area of security we are building a native Ceph encryption layer and want to enhance its security and functionality and evaluate the associated performance impact. Another example is in the area of reliability where we are investigating ability to do "live on-boarding" of volumes sitting externally to Ceph (e.g. volumes that were backed up to a cold storage like an object storage).
The work includes design, programming and performance and reliability evaluation. The work will be conducted in collaboration with the IBM Red Hat team and the Ceph open source community.
During the internship you will enhance your systems and storage skills, gain experience working with the open source community, and interact with both the IBM Research team and the Redhat engineering team.
Cloud Native control of data usage: Mesh for Data open source project
Managing enterprise data has always been a challenge. With the growth of cloud and hybrid-cloud adoption, data is stored and used in multiple environments, making access to data, security and governance become more challenging than ever.
Mesh for Data is a new open source project that aims to eliminates the complexity of proper and compliant data usage in Kubernetes clusters. It gets as input a description of a data centric workload along with non-functional requirements provided as policies defined by various personas such as a governance officer or operator. Mesh for Data translates the requirements into a deployment plan using an extensible library of modules that address the non-functional requirements. The deployment plan encapsulates the data user’s workload to ensure governance, secure access, and high performance when using data.
The internship is focused on adding features and functions that will drive how data usage is controlled. You will be experimenting with and developing code for open source projects, taking part in the Mesh for Data community. The work will be related to technologies such as containers, kuberenetes, service mesh, multi-cloud, workload orchestration, web assembly, Apache Arrow and more.
https://krb-sjobs.brassring.com/TGnewUI/Search/home/HomeWithPreLoad?Page...