The Research and Innovation organization in VMware’s Office of the CTO is driven by our pursuit of outcomes. We closely align and partner with our R&D business units to rapidly build differentiated technologies that often go from concept to product or service in less than two years. Last week at VMware Explore, we showcased several projects that we expect to significantly impact the VMware product and service portfolio in the coming months and years. Below are seven of those promising technologies I wanted to share:
1. Project Keswick
Edge environments often have diverse requirements and, in many cases, have limited to no IT staff to manage systems. This is why we see virtualization significantly enabling greater agility and velocity in edge environments. With our hypervisor, you can run VMs and containers along with network and security functions on a single device. Given that most public cloud services designed to run locally at the edge are packaged as containers, our platform has the potential to run them as well. Ultimately, that can allow organizations to change their applications and infrastructure at the edge simply via software updates, requiring no specialized IT skills physically present at those sites to facilitate the updates. This is where Project Keswick comes into play. With Keswick, you have a streamlined ESXi hypervisor that offers GitOps-style management. Infrastructure and application changes can be made by simply modifying an infrastructure or workloads definition YAML file and pushing the updates via GitHub. For a deeper dive into Project Keswick, see Simplifying Edge Deployments at Scale with VMware Project Keswick.
SplinterDB – the high-performance key value store invented by our Research team – is a core technology in vSAN’s new Express Storage Architecture. It provides high-performance metadata management. Beyond vSAN, we see SplinterDB as essential to many use cases because of its best-in-class performance and nimble codebase. Check out the blog post, Introducing SplinterDB: A High-Performance Key-Value Store, to learn more and contribute to the open-source project!
3. Project Trinidad
Project Trinidad leverages machine learning to fully model application behavior and detect anomalies. Beyond point-based anomalies, Project Trinidad can also model known good sequences in API calls – including header and payload – and alert and drive automation when suspicious behavior is detected. Customers and partners briefed on this project have been extremely excited about its far-reaching potential, and we are working hard to get this to market ASAP. To learn more and to join us as a design partner, see the post, Announcing Project Trinidad!
4. Federated Machine Learning
There is a robust future in federated machine learning technologies due to benefits such as privacy, data sovereignty, and reduced WAN/network bandwidth to transport large data sets to a cloud or data center. Our advanced development work has centered around two open source projects: FATE and OpenFL. Check out this post for insight into how we are incorporating OpenFL into Project Trinidad – Federated Learning With OpenFL for Microservices Applications.
We see this work as an important milestone because this marks the shift to a federating machine learning architecture in what we expect to be a future SaaS offering.
5. Project Newcastle
Cryptographic agility – the ability to decouple and manage cryptography independent of applications has garnered increased attention over the past year. While nearly all early crypto-agility conversations have focused on quantum readiness, we have been hearing from customers about their need for crypto-agility today. For starters, this takes away the heavy lift associated with refactoring applications to use stronger ciphers and is also of value when different use cases or local regulations require the same. At VMware Explore, we demonstrated how Project Newcastle integrates with Tanzu Service Mesh and what we believe to be the world’s first quantum-safe multi-cloud application. To learn more about the project and our VMware Explore technical demonstration, see Cryptographic Agility: A Deep Dive.
6. Project Narrows – Cloud Native Security Inspector
Project Narrows – developed by the same team of engineers that built Harbor – takes Harbor’s existing security capabilities even further. Project Narrows – recently open-sourced as Cloud Native Security Inspector – offers dynamic scanning that allows users to assess the security posture of containers at runtime. There’s already a large set of capabilities in the initial release, and we encourage the community to get involved and take Cloud Native Security Inspector even further. For a closer look, see Dynamic Security Scanning for Containers — Announcing Project Narrows.
WebAssembly (Wasm) is an open standard that defines a portable binary format for executable programs. Our Wasm Labs team has been busy building new developer tooling and proofs of concept that demonstrate the ultra-portability that can be achieved with Wasm. For example, at VMware Explore, we demonstrated running a modern website (running PreactJS, Vite, and MDX) using features such as server-side rendering – all running on a Raspberry Pi Zero. To learn more about our Wasm efforts, check out wasmlabs.dev, and to see more about our demo, including the code used to create it, see Modern websites in a Raspberry Pi Zero with WebAssembly.
Our deep collaboration that crosses VMware R&D and our customers and partners in the field has been essential to how we scope, execute, and continuously calibrate innovation projects against market needs. Please consider our opportunities for design partnership and look for more exciting innovations from us in the coming months!