IEEE Infrastructure Conference

Jennifer Fraser is the Sr Director of Infrastructure Engineering at Twitter, this includes Mission Critical Engineering, Site Operations, Supply Chain and Hardware Engineering. Jen is chair for @TwitterWomen, the employee resource group focused on gender equality. Jen manages the Twitter partnerships with @GirlsWhoCode and @techwomen to promote the engagement of women in STEM. Jen has traveled on delegations to Africa and the middle east to support women in STEM. Jen joined Twitter in 2012, immediately focusing on operational uptime of the physics supporting the platform to reduce service interruption.

Prior to joining Twitter Jen was the director of Engineering, Design and Construction for Vantage Data Centers, a Silverlake portfolio startup and has extensive experience in construction and operations. Jen has a Professional Engineering License in Mechanical Engineering and patents in Electrical transformation and storage. She is a founder/board member of a Mechanical Contracting company focused on infrastructure upgrades for operational efficiency and savings. Jen serves in her community on a local maintenance district directing tax revenue and prioritizing maintenance projects and contracts.

Dr. Danny Lange is Vice President of AI and Machine Learning at Unity Technologies where he leads multiple initiatives in the field of applied Artificial Intelligence. Unity is the creator of a flexible and high-performance end-to-end development platform used to create rich interactive 2D, 3D, VR and AR experiences. Previously, Danny was Head of Machine Learning at Uber, where he led the efforts to roll out a company-wide Machine Learning platform to support all parts of Uber’s business from the Uber App to self-driving cars. Before joining Uber, Danny was General Manager of Amazon Machine Learning where he also rolled out an internal company-wide Machine Learning platform used in all corners of Amazon. He also launched the first public Machine Learning cloud service on Amazon Web Services (AWS). Prior to Amazon, he was Principal Development Manager at Microsoft where he led a product team focused on large-scale Machine Learning for Big Data. Danny spent 8 years on voice-enable virtual assistants, first as CTO of General Magic, Inc., and then as founder of his own company, Vocomo Software. Danny started his career as a Computer Scientist at IBM Research.

Danny holds MS and Ph.D. degrees in Computer Science from the Technical University of Denmark. He is a member of ACM and IEEE Computer Society and has numerous patents to his credit.

From my first homemade computer to pushing the boundaries of Artificial Intelligence (AI) with the world’s most popular real-time 3D simulation engine, I have been fortunate to always have had the opportunity to work at the very forefront of computing infrastructure and its application. My work on the Unity platform blurs the lines between physical reality and the virtual. Not only does it reach 1.5 Billion people using Apps based on the Unity engine, but through massive cloud-based simulations we are able to train AI on scenarios that easily exceeds humanity’s collective experience in specific areas. Tracing the technologies I have worked on seems to depict a forward looking trajectory that carries a lot of promise in creating a better world for all, but it also raises some responsibilities that we as engineers need to take seriously. I see computer infrastructure rapidly spreading its tentacles deeper into the physical world while concurrently expanding into the virtual as well. The integration of large-scale computational and physical components enables new revolutionary scenarios ranging from self-driving cars over automated farming and healthcare to even scientific work itself. Cyber-physical systems with AI that eliminates the human-in-the-loop may soon take care of the world we live in. I see the most intriguing technical challenges ahead, but I also see significant issues in the area of ethics, security, reliability, and dependability of these systems.

Education: Stanford BS & MS Computer Science, UC Berkeley Haas MBA.

Industry history: Started programming at age 6, joined Oracle as a software engineer at age 17. Worked as a software engineer at Oracle & Versant; engineering management at E.piphany, SEVEN, and finally Google (2003-present). Currently responsible for Google Site Reliability Engineering (SRE), Networking, and Data Centers worldwide.

At small scale, both engineering and decision shortcuts are often rational because they are quick and easy to implement. With scale, some shortcuts become risky and expensive. We examine two common shortcuts which become less desirable as scale increases: Single Points of Failure, and Decision Heuristics incorporating Zero and Infinity.

Arya recently joined LightStep where he builds distributed tracing observability products and systems. Previously, he was the tech lead for the database team at Mixpanel where he built the column-store that powers all the real-time product analytics. Prior that, Arya was the tech lead for Twitter's timeline serving systems.

The focus of this panel is to discuss lessons learned, decisions and investment that each of the panelists have seen companies have incorporated when dealing with growth, scalability and reliability of their platform.

Raymond Blaine is the Director of the Cyber Research Center at the U.S. Military Academy (USMA) at West Point. In this capacity he is responsible for leading specialized research programs in support of USMA, Army, and DoD requirements as well as teaching cyber security subjects in the West Point curriculum. Immediately prior to this, he spent three years leading one of the twenty Army defensive cyber protection teams. He took his cyber protection team from inception through the traditional cybersecurity training and certification by U. S. Cyber Command. He then led his team through specialized training in industrial control system (ICS) security. His team became highly specialized in this area and was tasked with the defense of all Army ICS systems globally. These ICS systems reside on almost all Army installations and fall in to several of the Department of Homeland Security’s critical infrastructure sectors, to include dams, energy, and the defense industrial base.

Ray was among the first cohort of Army officers to be selected for the Army’s newest branch, Cyber Warfare. His military assignments include a variety of duty positions in the Cyber Protection Brigade at Fort Gordon as well as the 82nd Airborne Division and 18th Airborne Corps at Fort Bragg, NC. Ray was the Company Commander for HHC, 508th Parachute Infantry Regiment. He has served two tours in Operation Iraqi Freedom and one tour in Operation Enduring Freedom.

Ray received the BS from the U.S. Military Academy and the MS and Ph.D. degrees from Vanderbilt University, all in electrical engineering. His Ph.D. focused on radiation hardened by design analog and mixed-signal microelectronic circuits. His research interests are in industrial control system security, data visualization for cyber security, and reverse engineering. Ray is a member of the IEEE and ACM and is a licensed Professional Engineer in the Commonwealth of Virginia.

The United States Army is a microcosm of the country and has many, if not all of the same functions and challenges. These functions and challenges extend to cybersecurity and its roll in defending infrastructure. As the lead for one of the first cyber protection teams to attempt to tackle infrastructure security for the DoD I came across significant issues in regards to tools, personnel (team compositino, talent, retention, etc.), as well as typical organizational inertia. My team was able to overcome many of these challenges, but there are significant obstacles and opportunities for the larger community to tackle. The lessons and issues discussed in this talk are relevant to any large organization with a significant physical and logical footprint.

Nitzan Blouin has driven cultural changes in small startups and global companies. Today she is a Director of Engineering in Test at WeWork where she oversees test infrastructure, test engineering and development best practices. Before WeWork, Nitzan led a transformative effort at Spotify focused on big data quality by creating TC4D (Test Certified for Data) . She also bootstrapped a new team that focused on ML infrastructure and data validation. She has over 14 years of experience in building QA / Test departments and leading change in an organization’s quality mindset. Her daily meditation practice informs her simple, hands-on approach to solving complex problems.

Changing an engineering culture is one of the biggest challenges an engineering leader faces through her career; whether it’s scaling and hiring, delivering in a hyper-growth setting, infrastructure migration or adopting different processes and methodologies. Change is scary, even if it is for the best.

How do you get culture change done day in and day out when the finish line is years away? How do you justify the hefty price tag attached to your vision?

1. Here Comes The Sun: Be inspired by your vision. You are bringing to life a promised land where all the things will be solved. I dare say that the differentiating factor between success and failure depends on this step.
2. With A Little Help From My Friends: Gather the data. Surveys can sometimes be dry, but they are a great way to start a conversation, especially if your organization has hundreds of engineers. A survey will clarify your understanding of the problem; furthermore it will highlight the gap between how the organization perceives itself, and where it is at.
3. Come Together: You can’t do it alone. If part of your strategy includes hiring then here is your chance to bring in the best partners in crime who’ll execute your strategy. If you are not hiring, then find your allies in the existing organization. Better yet, do both.
4. Twist and Shout: Cultural change takes time, and often the closer you get to the finish line, the farther away it seems. Celebrate wins so they will keep you inspired and keep your vision alive.
5. Getting Better (All The Time) : The tactical implementation of your strategy will change from team to team, and over time, one size simply doesn’t fit all. Present the right context to the right audience. Account for changes in the organization, and in the tech stack, so your strategy is remains relevant.

Let’s make something clear - simple is not easy. Cultural change requires an understanding of a large organization in a short amount of time, pinpointing root causes and devising a compelling strategy that is not only tactical but also inspiring. Then there is execution: metrics, armwrestling for headcount, process changes, tools’ adoption, and so forth. The steps in this talk will demystify this complex process and you will take away actionable ways to change the culture in your own organization.

Dr. Tian-Ying Chang is a Sr. staff engineer and manager of Storage and Caching team in Pinterest. She leads the team to build the Storage and Caching infrastructure which powered several important Pinterest online systems, including Pinterest Homefeed, notification system, related pins, follower service, analytic system and so on. She is also leading the team to build a in-house time series database for metrics monitoring and general purpose analytic system. Before Pinterest, she worked in eBay and Microsoft on Operating System, Cloud, Hadoop/HBase. Tian-Ying holds PhD in distributed system from Georgia Institute of Technology and Bachelor degree in Computer Science from Peking University.

Goku is a highly scalable, cost-effective and high-performant online time series database service. It stores and serves massive amount of time series data without losing granularity. Goku can write tens of millions of data points per second and retrieve millions of data points within tens of milliseconds. It support high compression ratio, downsampling, interpolation, and multi-dimensional aggregation. It can be used in a wide range of monitoring tasks, including production safety, IoT, etc. It can also be used for real time analytics that make use of time series data.

Vinu Charanya is a Senior Software Engineer at Twitter building Twitter’s cloud infrastructure management platform. She is also a core team member of Women who code, a non-profit organization dedicated to inspiring women to excel in technology careers. She was also a part-time teacher and a mentor helping students learn iOS and Android development at CodePath. Vinu received her Masters in Computer Science and Engineering from the State University of New York.

Twitter serves over 320 million users in a month and at any given time processes over tens of millions tweets. It is powered by thousands of microservices that run on our internal Cloud platform which consists of a suite of multi-tenant platform services that offer Compute, Storage, Messaging, Monitoring, etc as a service. The multi-tenancy nature of these services makes it extremely difficult to track ownership of resources, effectively forecast capacity and meter resource utilization & cost. We built Kite, a service lifecycle management system aims to address these problems. In this talk, I’ll talk about how Kite as a system generalizes the service ownership tracking across Twitter’s disparate infrastructure & platform services. The ownership models are being extended to help manage resource provisioning, track operational metadata and credential management for services at Twitter. Kite has also enabled engineers @ Twitter to better visualize resources utilization and spend, plan capacity and empower them to make decisions around investing in features v/s infrastructure efficiency.

Berk is a Production Engineer at Facebook. He works at Traffic Foundation, focusing on the resiliency of Facebook's front end traffic delivery systems, one of the largest and busiest content delivery systems. He also contributes to FBOSS team focusing on Facebook's open source networking switch hardware and control plane software. Along with these he works on distributed systems resiliency in the infrastructure and site event (incident) review (postmortem) processes.

Previously he was a Site Reliability Engineer at Twitter working in various areas like Twitter API, web tiers, application and platform performance, and migration to Aurora & Mesos. Along with these he spearheaded complex incident response, production readiness reviews and postmortem processes; building Twitter's point-of-presences around the world; and founding Traffic Engineering team to work on network failure detection in the data centers and next generation load balancing & ingress traffic management.

Marius Eriksen is a Principal Engineer at GRAIL. Most recently, he has worked on parallel distributed computing and data engineering systems for genomic data processing and machine learning, including the design and development of the Reflow language and distributed runtime. Before GRAIL, Marius was a Principal Engineer at Twitter where he worked on systems infrastructure, including the design and implementation of Twitter's RPC system, Finagle.

Cloud computing offers powerful facilities that make for easy systems building: practitioners can solve a wide variety of problems by integrating existing infrastructure, with only a little application-specific code. The appeal of this approach is clear: we can build systems faster, using proven components that are usually operated by someone else.

However we should also be careful: When all we have is a cloud, then everything looks like it can be solved with 5 different services, 3 programming languages, some serverless computing, an object store, and an event stream. It’s all too easy to end up with systems that are far more complicated than they really should be.

In this talk, we’ll explore the design of Reflow, a cloud data processing system. Simplicity and parsimony are core to Reflow’s design, and we’ll talk about how we exploited these principles to create a powerful distributed data processing system, with a relatively small codebase and very few dependencies. While the talk is specifically about Reflow, most of its lessons are readily generalizable.

Sabrina Farmer is an Engineering Director in the Site Reliability Engineering org. She is the senior lead for the reliability teams for Google's largest products, as well as the various product infrastructure systems. Sabrina joined Google in 2005 and more than 20 years of experience in production engineering. She earned a bachelor's degree in Computer Science from the University of New Orleans, and is the founder of the “Scholarship for Women in Computer Science” at her alma mater.

Ihab Hamadi is Sr Director of Engineering at Western Digital, where he focuses on the innovation and growth of the company's platforms business, delivering data and application-centric solutions for the enterprise and cloud. He has a solid 21-year track record of building leading-edge cloud computing, data storage, networking, and virtualization technology and solutions. Before joining Western Digital, Ihab was Distinguished Technologist at HPE Aruba, leading development teams that contributed to OpenSwitch project and created ArubaOS-CX network operating systems, a modern OS for the mobile and IoT era, built from the ground-up to support Intent Based Networking. Ihab has also held a wide range of technology leadership positions with Broadcom, Emulex, and other companies. He earned an MS degree in computer engineering from the University of Denver.

The exponential growth in data is not only fueling new Big Data and Fast Data applications, it is also creating complexities in the way that data is being captured, preserved, accessed and transformed.

New NVMe™ over Fabrics (NVMe-oF) devices will soon be available from multiple vendors. These new fabric devices will require additional management procedures above and beyond what has been historically used.

The OpenFlex™ architecture disaggregates flash and disk, compute and network resources into independent, scalable pools that can be connected via common networking technologies, such as Ethernet. The open Kingfish™ API enables these pools to be presented as software composable infrastructure that can be quickly and easily orchestrated to precisely address the needs of complex and dynamic applications and data workflows. Various KingFish™ tools will also be contributed to open-source.

Amin Heydari is a Staff Capacity Engineer at Box Inc. where he is responsible for the capacity and performance planning of key components in boxes infrastructure. Over the past year, he led the effort for doing the capacity analysis for data migration from on-prem datacenter to public clouds. Prior to Box, Amin spend a few years at twitter, where he worked in various function, including Capacity Engineering, to help scale twitter during its hyper-growth phase.

Modern Enterprise Content Management Systems have unique challenges to securely store and manage access rights to content that is continuously growing at an exponential rate. Many of the cloud providers enable the ability to store large amount of content with the same relative costs as traditional solutions. The challenge is how to leverage these cloud providers to efficiently transfer and store content in multiple geographically distributed regions to met compliance requirements.

This presentation will cover capacity analysis modeling for content migration from datacenter to cloud providers while maintaining the necessary compliance requirements.

Ed is a distributed system expert, architect and developer of open source software. He is the co-author of TiDB, an open-source HTAP database, and TiKV, an open-source distributed transactional Key-Value database. Prior to PingCAP, Edward worked at Microsoft, NetEase Youdao and Wandou Labs.

The past ten years have witnessed the rapid development of the Internet and the popularity of distributed systems. A new field in distributed computing emerges: the convergence of the traditional relational databases and the distributed systems. In this talk, I will use TiDB as an illustration and introduce the trends and methodology of how to design modern databases that are both horizontally scalable and compatible with traditional relational databases.

Dinesh A. Joshi has been a professional Software Engineer for over a decade building highly scalable realtime Web Services and Distributed Streaming Data Processing Architectures serving over 1 billion devices. Dinesh is an active contributor to the Apache Cassandra codebase. He has a Masters degree in Computer Science (Distributed Systems & Databases) from Georgia Tech, Atlanta, USA.

Apache Cassandra 4.0 has several enhancements. One of the biggest enhancements is switching from blocking network IO using JDK Sockets to Non Blocking IO with Netty. As a result, Cassandra has seen gains in performance and efficiency. These gains translate into real world costs and allow Cassandra to scale better. This presentation will take you on a tour of the improvements of Cassandra's network layer (old & new) and help quantify the gains in real world terms.

Dr. Bo Liu is currently a software engineering manager at Pinterest. His team is responsible for building online stateful high performance distributed systems such as unified ML serving platform, real-time big data analytics system with online updates, HDFS based read-only KV serving system and a set of RocksDB based sharded and real-time replicated systems. These systems serve tens of millions of requests per second and power majority of Pinterest products. Bo Liu is the author of Pinterest core C++ library which helps developers solve common difficulties of building large-scale stateful services, such as real-time data replication, cluster management, low level client-side library, resilient request routing, stats reporting, etc. Before Pinterest, Bo Liu worked on TAO as a software engineer at Facebook. TAO powers the entire Facebook social graph and is arguably the largest distributed data system on the planet in terms of QPS. Bo Liu spent a short period of time in academia after his PhD.

In the past three years, Pinterest has evolved from a state where the majority of the product surfaces were powered by offline daily jobs to being fully dynamic and responsive. During the same period of time, the amount of the system traffic also greatly increased due to a combination of user growth and more complex products. To make our systems meet the requirements by these changes, we had to evolve the architecture of existing services, better leverage solutions provided by AWS, and build nearly ten brand new services. In this talk, we will discuss how Pinterest backend systems evolved in the past three years to power fully dynamic and responsive products. An overall picture of the current backend systems will be shared. We believe some of the systems solve common problems of distributing personalized content in a dynamic and responsive way. Technical details about these systems will be discussed. We will also share the experience we learned about how to better leverage AWS solutions to build more efficient and reliable backend systems.

Norman Maurer is a Senior Software Engineer @ Apple, where he works on network frameworks / libraries that are used to power mission critical production systems. He is the project lead of Netty, which is one of the most used (if not the most used) network frameworks for the JVM. 2017 he was also elected as a Java Champion. Beside his work on JVM based libraries he also works on SwiftNIO, which was open sourced earlier this year and is a port of Netty to Swift.

Netty is the underpinning for most network frameworks for the JVM, or is used directly by services which provide network connectivity and run on the JVM. In many of these cases, including Apple, Netty needs to operate at a massive scale and plays a critical role in the service's infrastructure.

To operate Netty based services at Apple's scale we heavily invest in the development of Netty. Some of the changes we made over the last years dramatically improved the performance and lowered the latency of Netty based solutions including Cassandra and web frameworks built on top of Netty.

This talk will demonstrate different techniques we use in Netty to achieve low latency and high throughput and review how we use Netty to scale our services and therefore our infrastructure; including what worked and lessons learned while making performance and scalability improvements.

Stefan Nastic, PhD is a co-founder and CEO of Reinvent. He also works as a Postdoctoral Research Assistant at the Vienna University of Technology (TU Wien), Austria. His main areas of expertise include: Internet of Things and Edge Computing; Cloud Computing; and Blockchain Technologies. Over the last decade, he has been involved with a number of large industrial and research projects related to Big Data, Internet of Things and Cloud Computing, such as a multi-million Pacific Controls Cloud Computing Lab (PC3L). During this time, Stefan has gained an extensive experience acting as project manager/lead, product owner and software architect. During the lasts couple of years, his main focus was on advancing various aspects of Distributed Ledger Technologies (i.e. Blockchain) – ranging from the development of decentralized applications, designing hybrid-computing architectures to transforming traditional business models with decentralized trust management.

Over the last decade, sharing economy has reached a global scale. Recent technological advances have enabled widening of the "circle of trust" by increasing the volume and geographical scale of possible interactions among the participants of the sharing economy. Unfortunately, current implementations of sharing economy platforms have several limiting factors, the most significant being the centralization of trust. The centralized trust goes against the inherently decentralized nature of the sharing economy, hence limiting the possibilities for a wider adoption, beyond specific domain verticals. In this talk, we provide an overview of the WeValue platform, which is currently being developed by Reinvent – a Vienna University of Technology spin-off company. WeValue is a facilitating platform for the sharing economy that provides the concept of decentralized trust at its core. We present the most important features of the WeValue platform and focus on the core technical challenges of building a hybrid-computing platform, which combines traditional centralized computing infrastructures with the Blockchain-based decentralized computing infrastructure. Finally, we outline a particular use case of utilizing the WeValue approach to realize resource and data sharing for IoT systems.

Ruben has been working on Twitter’s RPC stack for the past five years. In that time, he has made substantial contributions to both the design and implementation of Finagle (https://github.com/twitter/finagle/) which have markedly improved the resiliency and operability of Twitter services.

Twitter's RPC framework (Finagle) employs non-cooperative, client-side load balancing. That is, clients make load balancing decisions independently. Although this architecture has served Twitter well, it also has its drawbacks. In particular, it results in a mesh network where clients open connections to all backend servers in order to maintain a globally equitable load distribution. This scales poorly with larger clusters for a number of reasons. In this talk, we will dive deeper into the problem space and how we mitigate some of the drawbacks via an algorithm we call "Deterministic Aperture".

Keith Packard has been developing system software since 1986, working on Unix, the X Window System and Linux. He is currently a Distinguished Technologist at Hewlett Packard Labs working on memory semantic fabrics and memory driven computing. Keith received a Usenix Lifetime Achievement award in 1999, an O'Reilly Open Source award in 2011 and sits on the X.org foundation board.

Gen-Z is an open systems Interconnect designed to provide memory semantic access to data and devices via direct-attached, switched or fabric topologies. Support for this will require additional kernel infrastructure, device drivers and management systems. This presentation will first briefly describe the Gen-Z interconnect, outline the management required for Gen-Z to operate and finally present an architecture for supporting Gen-Z natively in Linux.

Prasanth Pulavarthi is Project Manager on the Microsoft AI Platform team where he is responsible for AI/ML training and inferencing frameworks. He drives Microsoft’s participation in ONNX (Open Neural Network Exchange), the open standard for interoperable AI models. The ONNX format provides a common format that helps bridge the gap between data scientists/researchers creating models and developer deploying models at scale. Previously Prasanth has worked on a variety of large scale, globally distributed infrastructure projects including Bing, OneDrive, and Advertising.

Traditionally each machine learning framework has been its own silo. Data scientists needed to create, train, and deploy models with the framework and its proprietary tools. Many times, deploying with the framework was not practical and much development effort had to be spent to convert it to another framework or format. To remedy this situation, Microsoft co-created Open Neural Network Exchange (ONNX). ONNX provides a common format that enables frameworks (Caffe2, Chainer, CNTK, MXNet, PyTorch, SciKitLearn, TensorFlow) to produce models that can be deployed in a variety of devices and platforms (Azure, Windows, Linux, iOS/CoreML, Android, etc.) ONNX is an open industry effort with participation from many companies in the industry.

We will share the motivations for a cross-industry partnership, the current technical state and roadmap for ONNX, how ONNX is supported at Microsoft, and show how to go from training to deployment using ONNX models.

Dr. Peter Sbarski is VP of Engineering at A Cloud Guru, AWS Community Hero, and the organizer of Serverlessconf, the world’s first conference dedicated entirely to serverless architectures and technologies. His work at A Cloud Guru allows him to work with, talk and write about cloud computing and serverless architectures. He is the author of Serverless Architectures on AWS (Manning Publications, 2017) and is currently collaborating on another book called Serverless Design Patterns with Tim Wagner and Yochay Kiriaty.

Peter is always happy to talk about cloud computing and can be found at conferences and meetups throughout the year. Peter’s passions include serverless technologies, event-driven programming, back end architecture, microservices, and orchestration of systems. Peter holds a PhD in Computer Science from Monash University, Australia.

A Cloud Guru built one of the largest platforms using serverless technologies like AWS Lambda (we don’t run a single server, anywhere). It has been an incredible journey and we learnt a lot of lessons scaling our platform to 600,000+ users.

In this session we’ll share notes from our experience including:

• What our serverless system looks like 2 years in (including key metrics)
• Important design patterns and architectures
• Common serverless mistakes and how to avoid them

We will dive in to the design of our platform and share interesting data, go through key patterns and architectures, and discuss what it actually takes to build a scalable, reliable and high-performing cloud-native serverless systems today.

Randy Shoup has been building and operating service-oriented distributed systems for more than 20 years. He was Chief Engineer at eBay, led engineering for Google App Engine, and was VP of Engineering at Stitch Fix. He is currently VP of Engineering at WeWork in San Francisco.

Over time, almost all large, well-known web sites have evolved their architectures from an early monolithic application to a loosely-coupled ecosystem of polyglot microservices. This presentation discusses what a healthy microservice ecosystem looks and feels like to the service owner, based on the speaker’s experience building and operating services at Google and eBay.

The presentation starts with the considerations in designing services to interact effectively with clients and dependencies, including interface stability in the face of evolving functionality, as well as some patterns and anti-patterns for service design.

It next describes the service ecosystem itself - how service teams relate to one another, how to encourage -- but not enforce -- standardization, and how to grow an ecosystem without central control.

It concludes by discussing operating services at scale, including the challenges of zero-downtime deployments and data migration.

Sandy is the SRE Tech Lead for the Revenue Platform at Twitter. Over the years she has worked in software engineering and operations roles at a variety of companies, including two startups and a Fortune 500 company. Her career has spanned different segments of the industry including video games, digital asset management, web hosting, and social media. She has been a community organizer in LA and SF, in areas including Python, Django, and SRE. She was a founding board member of PyLadies.

The Ad Server is Twitter’s revenue engine. It performs ad matching, scoring, and serving at an immense scale. The goal for our ads serving system is to serve queries at Twitter-scale without buckling under load spikes, find the best possible ad for every query, and utilize our resources optimally at all times. This talk will provide an overview of capacity planning strategies and lessons learned that have proven to be valuable in achieving these goals.

Jamie Turner is the Director of Persistent Systems Engineering at Dropbox, where he leads the software platforms that store Dropbox customers’ data. These teams collectively build and run systems which hold many trillions of metadata records and exabytes of file data.

Prior to this management role, he was a Principal Engineer, performing various design, implementation, and technical leadership roles on Magic Pocket, Dropbox's block storage system. Additionally, he started an initiative to significantly reengineer Dropbox’s client-side sync engine.

Before Dropbox, Jamie was Head of Engineering at Bump Technologies, Inc--a mobile startup acquired by Google in 2013. Prior to Bump, he was VP of Information Services at YouGov, plc. In addition to industry work, he's authored several open source projects that have accumulated thousands of stars on GitHub, including the Stud SSL proxy.

Over the past five years, Dropbox has significantly transformed its business by moving its primary multi-exabyte storage system from 3rd party cloud providers to in-house infrastructure and software.

Jamie will first outline the history of this in-house project, from inception and early development, through a massive migration, and then on to stability and maturity.

Then, the talk will dive deeper into some aggressive improvements pursued in the last two years that have enabled the system to achieve world class durability, performance, and cost efficiency.

Dathan Vance Pattishall, a Principal Engineer in Service Infrastructure for WeWork and has been working in the Valley since 1994.

Delivering infrastructure to engineers requires more than just great engineering. WeWork’s Service Infrastructure team embraced product-focused processes in order to provide a fantastic Kubernetes experience that engineers love. Learn about how we brought cutting-edge deployment practices to the whole organization by interviewing stakeholders, communicating the pain points to team members, using a crawl-walk-run approach, and the tradeoffs of delivering an MVP instead of everything at once.

Natalia Vassilieva is a Sr. Research Manager at Hewlett Packard Labs. Her focus is machine learning and artificial intelligence, data analytics, and application-driven software-hardware optimization and co-design. She leads the Software and AI group which is currently working on performance characterization and modelling of deep learning workloads, fast Monte Carlo simulations, and systems software, programming paradigms, algorithms and applications for Memory Driven Computing – a new concept of computer architecture by Hewlett Packard Enterprise. Natalia joined HPE at 2007 and served as the head of HP Labs Russia from 2011 till 2015. Prior to joining HPE, she was an Associate Professor at St. Petersburg State University and worked as a software engineer for different IT companies. Natalia holds PhD in computer science from St. Petersburg State University.

There is a vibrant and fast growing ecosystem of software and hardware for deep learning. Various deep learning frameworks are available for anyone who wants to try out this technology for their needs. With the variety of choices in hardware configurations and software packages, it is hard to pick the most optimal tools. The effectiveness of any hardware/software environment varies depending on a deep learning workload. HPE Deep Learning Cookbook is a set of tools to guide the choice of the most optimal hardware/software configurations for any given deep learning application. It is based on extensive benchmarking of reference workloads and performance modelling. The talk will present two main components of the cookbook toolset and general learnings on infrastructure requirements and performance of deep learning applications.

Lohit is part of Hadoop and Log Management team at Twitter. He has been concentrating on scaling Hadoop FileSystem, Hadoop Resource Manager, Log Ingestion and Processing pipelines at Twitter. Previously he has worked at few startups building scalable file systems and was also part of Hadoop team at Yahoo! when it was open sourced. He has Masters degree in Computer Science from Stony Brook University where he was guided by Professor Steven Skiena.

Log files consisting of events from different services are a rich source of information for large scale analytics. Events can be as simple as log line or as complex as nested structured objects like thrift or protobuffers. At Twitter every service generates events within a particular “category” and publishes them to the Event Log Management framework. The framework aggregates events from the same category into log files, usually stored on a distributed file system like the Hadoop Distributed File System (HDFS). In this presentation we provide an overview of the Event Log Management framework used at Twitter, highlight its advantages, and compare it with similar frameworks. Twitter services generate trillions of events with aggregate size exceeding multiple petabytes of data every day. The main focus of our efforts has been efficient use of hardware resources, scalability and reliability of the framework. We present our framework as a set of modular components which has scaled for event log management at Twitter.

Zhenzhao works at Twitter as part of Hadoop and Log Management team. He is currently concentrating on Twitter Log Ingestion Pipeline which scales to handle trillions of events per day. Previously he was a member of Distributed File System(Pangu) team in Alibaba Cloud where he focused on random file access in Pangu used as storage for Virtual Machines. He has Bachelor’s degree from Nankai University and Master’s degree from Tsinghua University.

Andy is a Site Reliability Engineer in the Infrastructure Optimization and Performance team at Twitter. Formerly he worked at Lab126 (Amazon) providing low-level optimization and tools for the Kindle Fire Android platform. He was CTO of Integratus, a high-availability and file-system startup that was sold to Veritas Software. Andy holds computer science degrees from Purdue University (BS) and University of Florida (MS).

The venerable zlib library has been the workhorse of general purpose compression for nearly three decades, offering reasonable throughput and compression ratio. Recent advances have led to new compression codecs with improved characteristics. We discuss the applications of ZStandard (Facebook) and Brotli (Google) at Twitter. We also discuss our experience with optimization of zlib and LZ4 via vectorization.