AWS re:Invent 2017

Many financial institutions want to provide greater autonomy to their developers but find themselves hemmed in by intricate legacy processes and centralized IT teams that own different parts of the infrastructure and application stack. For Barclays PLC, the solution to this impasse lay in automating the management of hundreds (and, soon, thousands) of accounts, granting developers access to AWS technology directly via AWS Console or API while conforming to bank policy. Key to Barclays' strategy is “Persephone,” a system that creates accounts, sets policy, defines which services are enabled in the account, and runs Lambda functions to ensure continuous compliance. Persephone relies on services including CloudTrail, CloudWatch Events, IAM, KMS, and others to balance control and productivity. Barclays is also working to introduce automated reasoning testing and assurance techniques made available by the Automated Reasoning Group at AWS. Attendees of this session will gain an understanding of the mindset, skillset, and toolset shifts that a diversified global financial institution must make to meet the control objectives of the present while ensuring the agility to accommodate technology change of the future.

The Bank of Nova Scotia is using deep learning to improve the way it manages payments collections for its millions of credit card customers. In this session, we will show how the Bank of Nova Scotia leveraged Amazon EC2 Container Service and EC2 Container Registry and Docker to streamline their deployment pipeline. We will also cover how the bank used AWS IAM and Amazon S3 for asset management and security, as well as AWS GPU accelerated instances and TensorFlow to develop a retail risk model. We will conclude the session by examining how the Bank of Nova Scotia was able to dramatically cut costs in comparison to on-premise development.

How do you get your security and compliance team to embrace the cloud? "Getting to Yes" with Vanguard's Security, Legal, and Compliance Teams was a key factor to the organization's journey to the cloud. Maintaining a high level of assurance is solvable when using an iterative, agile approach. Vanguard is taking existing on-premises controls, plus cloud frameworks such as NIST, CSA, etc., to develop the right set of cloud controls that provide maximum security without sacrificing business agility. In this session, we cover: Vanguard's approach to developing appropriate controls for its cloud deployments; key considerations and best practices when implementing controls; leveraging the AWS Cloud Adoption Framework and the four security perspectives to map controls appropriately; and the various AWS services (IAM, Amazon VPC, AWS KMS, and AWS CloudTrail) that we leveraged. We also cover the iterative and agile approach we are taking by embracing DevSecOps principles.

FINRA's analytics platform unlocks the value in capital markets data by accelerating trade analytics and providing a foundation for machine learning at scale. The platform enables FINRA's analysts to perform discovery on petabytes of trade data to identify instances of potential fraud, market manipulation, and insider trading. By centralizing all data in S3, FINRA's architecture offers improved agility, scalability, and cost effectiveness. Analytics services such as Amazon EMR and Amazon Redshift have freed FINRA's data scientists from the constraints of desktop tools, allowing them to apply machine learning techniques to develop and test new surveillance patterns. All of this is done while meeting FINRA's security and compliance responsibilities as a financial regulator. At the end of this session, you'll have an understanding of how to apply FINRA's architecture to trade analytics and other financial services use cases, including meeting regulatory requirements such as the Consolidated Audit Trail (CAT) reporting.

Many enterprises that follow regulated, process-driven workflows would like to take advantage of the innate features and benefits of AWS to become more agile, achieve operational excellence, and accelerate time-to-market while leveraging a DevOps culture and development methodology. But building a mature DevOps capability doesn't happen overnight. Creating and implementing testing, compliance, and security automation frameworks requires time and organizational and process changes. Financial institutions are addressing this challenge by using AWS Service Catalog to help bridge the gap between traditional operations and true DevOps.

For years, Riot Games deployed to their own private data centers across the globe to meet the growing demands of their game, League of Legends. The last seven years saw explosive growth in new data centers worldwide, along with a great deal of technical debt. This is Riot Games' story of how they overcame their technical debt by taking the League of Legends platform into the AWS Cloud. AWS services gave Riot Games the infrastructure agility they were lacking within their data centers and empowered them to focus more on new player features and less on legacy infrastructure. In this session, you learn why and how Riot Games migrated their existing platform to the AWS Cloud and the advantages gained by the move using services such as Amazon EC2, Elastic Load Balancing (with Application Load Balancers), Amazon EBS, and Auto Scaling. Riot Games also shares how they created new automation toolsets to enable the existing tools and replace a few legacy ones.

Learn how Gearbox integrated Amazon GameLift into "Spark", the cloud-ready infrastructure that powers all of their games. In this session, Gearbox talks about the mental shifts they had to make when moving their games online, as well as the unexpected challenges. Gearbox also dives deep into their experience of integrating the Amazon GameLift Server SDK, running multiple processes with different parameters, asynchronous build deployment, global fleet management, load balancing, and scaling to meet unexpected player demand. You'll learn Amazon GameLift best practices that can help simplify game session management, reduce engineering overhead, and optimize for player experience.

Given the complexity and scale of contemporary games, it's been a dream of game creators to algorithmically generate game content. Nexon wanted to create a large-scale, open-world MMORPG called Durango, where algorithmic generation is desperately needed to minimize development costs and maintain game longevity. In-game objects such as trees and plants are placed based on complex rules, with the intention of mimicking the realistic ecosystem that evolves continuously. However, a large amount of computation and a careful orchestration of various computing resources are required due to the immense size of the in-game lands. Nexon achieved this goal by leveraging AWS services to take advantage of massive parallelism supported by the infrastructure. In this talk, Nexon discusses the architecture they settled on for algorithmic generation of game content in a large scale, and the AWS services involved such as Amazon SQS and Amazon ECS with automatic scaling and spot instances.

Learn how Rovio replaced functionalities in their analytics platform provided by a 3rd party vendor in seven weeks with Amazon Athena, Amazon EMR, and Amazon Redshift. In the ever changing world of games, you must stay ahead of trends and understand your customer base's actions. We describe how we built a data lake using AWS services such as Amazon S3, Amazon Athena, and Amazon EMR. We give a general overview on our analytics stack, business problems unique to the gaming industry we're trying to solve, data philosophy (one truth, open data), and challenges we had at the beginning of this year. We also dive deep into areas most gaming companies struggle with, like how we manage the schemas, convert data to efficient columnar format, partition the data, ultimately use the data, and handle some of the challenges we faced using services like Athena. Learn about our quest to perfect games and customer gaming experience!

This session covers how the team at Ubisoft evolved For Honor's infrastructure using Amazon ECS and supporting systems (Amazon CloudFront, Amazon ElastiCache, Amazon Elasticsearch Service, Amazon SQS, and AWS Lambda, with monitoring through DataDog) from a proof of concept to an infrastructure as code solution. The team shares war stories about supporting both internal and live environments, and the challenges of bridging cloud and on-premises systems.