Nov 2014 – Aug 2019 · 4 yrs 9 mos · Palo Alto, CA

• I used to run engineering, go-to-market, product management, and operations for multiple managed MySQL-family relational database services including Aurora, MySQL, and MariaDB.
• Aurora is the fastest growing service in Amazon Web Services (AWS) history and has experienced phenomenal growth in the last 3 years. MySQL is the oldest and largest managed relational database service that is offered by AWS. We develop the novel Aurora database engine as well as a control plane that is shared across the different services. In addition we are also responsible for software deployments, operations, support and customer experience starting from how the services are enhanced, supporting new instance types and engine versions, to on-boarding customers and ensuring their success.
• Amazon Aurora (see our recent SIGMOD 2017 paper) is a managed relational database service for OLTP workloads that is designed around the notion that the central constraint in high throughput data processing has moved from compute and storage to the network. Aurora brings a novel architecture to the relational database to address this constraint, most notably by pushing redo processing to a multi-tenant scale-out storage service, purpose-built for Aurora.
• Most recently we announced the impending availability of three major evolutions of the Aurora service. The first is a multi-master version of Aurora that will support multiple writers in a cluster. The second is the “Serverless edition” of Aurora that is designed for workloads that are highly variable and subject to rapid change where customers only need to pay for the database resources they use, on a second-by-second basis. The third is support for light-weight analytical/reporting queries by scaling out the query processor into the storage tier.

May 2012 – Oct 2014 · 2 yrs 5 mos · San Francisco Bay Area

• After the acquisition of Truviso by Cisco, I was responsible for setting and driving the overall technical strategy for Big Data and Analytics for Prime Analytics – the new Cisco product that embodied the Truviso technology. In this capacity, I provided technical leadership for multiple teams that built the core platform of Prime Analytics (based on the original Truviso product) as well as vertical solutions relevant to Cisco’s business. Examples of these vertical solutions include: WiFi Location and Reporting Analytics (for high volume WiFi deployment), Service Assurance for Cloud Orchestration, Retail Analytics for sensor data in stores. In addition to these business focused solutions, I have been involved in research efforts devoted to pushing analytics and data processing to the edge of the network where we have successfully embedded the Truviso streaming engine in the network elements themselves (i.e., containers that run in switches and routers).

Jan 2006 – May 2012 · 6 yrs 4 mos · San Francisco Bay Area

• We sold Truviso to Cisco Systems on May 3, 2012.
• Truviso is an innovative startup that pioneered a real-time analytics solution for high-volume and high- velocity Big Data. I co-founded Truviso and led the commercialization of its core technology that was based on groundbreaking work from the UC Berkeley Database Systems research group where I did my Ph.D. The key technology underpinning the Truviso engine is to continuously process high-volumes of streaming data on-the- fly before it even hits disk. This revolutionary approach of stream data processing was implemented in an evolutionary package by extending the open-source PostgreSQL database system to process streams as well as traditional relations. At Truviso, I was responsible for setting and driving the overall technical strategy and direction for the Truviso product and solution portfolio. In addition, I worked in close collaboration with marketing, sales and engineering teams in managing the product and solution roadmap, performance engineering, and technology evangelism. Previously, I built and managed the initial engineering, services and support teams at Truviso. In these various capacities, I have:
• Consummated an M&A transaction with Cisco Systems
• Participated in raising multiple rounds of financing for the company.
• Evangelized and improved the techniques I investigated in my doctoral research on stream processing.
• Architected, designed and implemented the Truviso product with the first GA release and production
• customer in 11 months.
• Led the effort in identifying and securing intellectual property in the form of patents and trade secrets.
• Hired an exceptional team of A+ engineers and managers – our team continues deliver strong releases of
• the product. All hires were through the team’s network.
• Built out the whole product including tools and connectors to various platforms by heavily leveraging open
• source software components produced for and as part of the PostgreSQL project.

Sep 2001 – Dec 2005 · 4 yrs 3 mos

• Research on data stream management. Ph.D. Dissertation: Shared Query Processing in Data Streaming Systems.
• More information at: http://www.cs.berkeley.edu/~sailesh

Jun 2000 – Sep 2001 · 1 yr 3 mos

• Research and development work at the Database Technology Institute:
• 1. Memory-centric database research. This effort investigated efficient database system implementations where entire tables and indexes can remain resident in memory.
• 2. Database caching for internet applications as part of the DBCache project.

May 1998 – May 2000 · 2 yrs

• R&D of database systems: I worked as part of, and then led, a team that built Object-Relational technology in the DB2 UDB product. My contributions:
• 1. Delivered ADTs or structured column types (with inheritance and methods) for DB2. This involved teaching almost every major component (e.g., executor and storage-manager) in the DB2 engine the use and behavior of structured types.
• 2. Developed approaches for binding database ADTs to objects in programming languages (in client applications and server-side methods/functions) thereby influencing the JDBC 2.0 and SQLJ standards.
• 3. Designed an approach to exchange ADTs between database servers through an open standard
• 4. Led the design work for the support of ADTs in tools such as Export, Import, Load, Replicator.
• 5. Influenced the design of the SQL99 standard by working closely with Standards reps to ensure the coherence of the language and the ability of IBM's DB2 products to conform to the standard.

Jul 1997 – Apr 1998 · 9 mos

• I worked on various aspects of Java virtual machine development at Netscape. This included work on Netscape's legacy VM as well as a next generation effort called Electrical Fire. Specifically, I accomplished:
• 1. Developing Java Native Interface (JNI) assembly language invokers for various platforms.
• 2. Working on runtime support for Electrical Fire (a high performance Java JIT compiler).

May 1996 – Aug 1996 · 3 mos

• Extended trace file formats for ATAC: a coverage analysis tool now called xSUDS. I worked with a team deploying ATAC - a coverage analysis tool from Bellcore and IBM - for use with developers of the DB2 UDB product. My contribution was to extend the trace file formats of ATAC for more efficient storage of information.

Aug 1995 – Jul 1997 · 1 yr 11 mos

• Performed research on software testing and reliability, using tools based on code-coverage. My master's thesis explored the problem of using code-coverage to quantify the reliability of large software systems.