Top 10 Enterprise NoSQL Use Cases | 2 The Top 10 Enterprise NoSQL Use Cases Table of Contents Executive Summary......................... Introduction....................................... Personalization................................. Profile Management........................ Real-Time Big Data.......................... Content Management.................... Catalog................................................ Customer 360° View....................... Mobile Applications........................ Internet of Things............................ Digital Communication.................. Fraud Detection................................ Conclusion......................................... Table of Contents 3 4 6 7 8 9 10 11 12 13 14 15 16 Top 10 Enterprise NoSQL Use Cases | 3 NoSQL powers revenue-driving, customer-facing applications Major enterprises are deploying NoSQL for customer-facing, revenue-driving applications that serve hundreds of millions of consumers, business customers, and citizens. They’re turning to NoSQL to overcome limitations of decades-old relational database technology. And they’re choosing open source NoSQL over proprietary relational products from providers like Oracle, IBM, and Microsoft – in many cases, replacing those databases with NoSQL – at a fraction of the cost. Top 10 use cases This document provides a guide to the top 10 enterprise NoSQL use cases we’re seeing in the market today. We based the list partly on formal survey data (taken in Q4 2014) and partly on current conversations with customers and prospects. The top 10 are: Key takeaways For both business and technology decision makers, there are four key takeaways from this document: 1. NoSQL adoption has moved beyond experimenters and innovators to mainstream “early adopters” for mission critical applications 2. NoSQL is being adopted by major companies in virtually every industry for a broad and expanding set of use cases 3. Early adopters are realizing significant, quantifiable benefits and first-mover advantages: improved application performance, easier and more affordable scalability, faster development cycles, and more efficient resource utilization 4. NoSQL is now a fundamental cornerstone of modern, big data infrastructure Executive Summary NoSQL, now ready for prime time Over the last few years, NoSQL database technology has experienced explosive growth and accelerating use by large enterprises for mission critical applications. For example: Tesco uses NoSQL to support its catalog, pricing, inventory, and coupon applications McGraw-Hill uses NoSQL to power its online learning platform Sky uses NoSQL to manage user profiles for 20 million subscribers Neiman Marcus uses NoSQL to enable personalized customer engagement across all channels Sabre uses NoSQL to support one of the world’s largest travel data services KDDI, Japan’s #2 telecom with 20 million subscribers, uses NoSQL to manage its cloud services ID platform The Turkish Ministry of Health uses NoSQL to power a national health database serving 77 million citizens Ryanair turns to NoSQL to deliver a great experience for its mobile app users From Internet to Enterprise What’s remarkable about that list – which are just a few examples from hundreds of enterprises deploying NoSQL – is the variety of organizations and industries represented. Eight to 10 years ago, when NoSQL pioneers first deployed the technology, its use was limited to Internet-age companies like Google, Amazon, Facebook, and LinkedIn. Today, enterprises across virtually every industry are deploying NoSQL. In short, NoSQL is now ready for enterprise prime time. Personalization Profile Management Real Time Big Data Content Management Catalog Customer 360° View Mobile Applications Internet of Things Digital Communication Fraud Detection Executive Summary Top 10 Enterprise NoSQL Use Cases | 4 Introduction: A New Era of Enterprise Database Infrastructure What’s Driving NoSQL: Mega-Trends and New Business Requirements According to nearly 90% of respondents to a recent customer survey (Q4 2014), NoSQL is “important” (56%) or “critical” (33%) to their business. The survey represented a broad set of industries: Advertising and Marketing Research, Automotive, Financial Services, Gaming, Manufacturing, Media, Retail, NoSQL adoption is being driven by both industry trends and business objectives. Major trends – the proliferation of mobile devices, the Internet of Things, big data, and cloud infrastructure – are raising the scalability and performance requirements for enterprise web and mobile applications, as the number of users, user expectations, and user interactions continue to grow. Security, Software and Technology, Telecommunications, and Travel. NoSQL has been deployed in production for one to two years by 47% of them, over two years by 40%. Top benefits of NoSQL include improved application performance and availability as well as improved resource and operational efficiency. NoSQL is enabling enterprises to increase business agility, achieve a faster time to market, and operate at a global scale while reducing costs. In addition, enterprises are modernizing database infrastructure by implementing database as a service (DBaaS) to meet the requirements of individual business units with and without multi-tenant cloud infrastructure, public or private, to improve resource and operational efficiency. Introduction How Important is NoSQL? Critical Important Neutral 38% 53% 9% Is NoSQL Deployed in Production? Yes, 1 - 2 years Yes, 2 - 3 years Yes, 3+ years 47% 24% 18% No12% How Important is NoSQL? Critical Important Neutral 38% 53% 9% Is NoSQL Deployed in Production? Yes, 1 - 2 years Yes, 2 - 3 years Yes, 3+ years 47% 24% 18% No12% How Did NoSQL Benet the Business? Improved Application Performance Increased Availability Increased Resource Eciency Improved Customer Experience Increased Operational Eciency 85% 53% 50% 29% 21% Why Was NoSQL Required? Better Performance Scale Out Agile Development Reduced Downtime Flexible Data Modeling 69% 69% 50% 47% 44% Multiple Data Centers 28% Decreased Costs 22% Cloud Deployment 16% Top 10 Enterprise NoSQL Use Cases | 5Introduction Overcoming the Challenges – and Costs – of Relational Databases These industry trends and business requirements present new challenges for traditional relational databases. Relational databases are difficult to scale, and they’re unable to guarantee the performance and availability necessary to meet increasing user expectations in the Internet age of always-on business. Whereas relational databases are limited by their “scale up” architecture – i.e., increasing the processing and storage capacity of a single server to support more users and operations – a NoSQL distributed database like Couchbase Server employs a “scale out” architecture that enables easy, affordable, on-demand scaling simply by increasing the number of servers. How Many Operations/sec on NoSQL? Less than 50K 50K - 300K 301K - 500K 41% 41% 9% 501K - 1M9% How Much Data is Stored in NoSQL? 1 - 10TB 200GB - 1TB Less than 200GB 21% 52% 21% More than 10TB6% How Many Operations/sec on NoSQL? Less than 50K 50K - 300K 301K - 500K 41% 41% 9% 501K - 1M9% How Much Data is Stored in NoSQL? 1 - 10TB 200GB - 1TB Less than 200GB 21% 52% 21% More than 10TB6% Mastering big data is a fundamental requirement for every large enterprise. NoSQL and Hadoop have become cornerstones of modern big data infrastructure. They are complementary technologies. While the Hadoop ecosystem is engineered for offline analytical workloads, NoSQL is For a more complete discussion of what’s driving enterprise adoption of NoSQL – and how NoSQL overcomes the limitations of relational engineered for online operational workloads. By integrating NoSQL and Hadoop – along with high-speed messaging tools like Kafka and stream-based analytics like Storm and Spark – companies can meet both analytical and operational requirements for a real-time big data solution. database technology – download our white paper, “Why NoSQL?” at http://www.couchbase.com/nosql-resources/what-is-no-sql. NoSQL and Hadoop: Cornerstones of Modern Big Data infrastructure Building Blocks of Modern Big Data Infrastructure Top 10 Enterprise NoSQL Use Cases | 6 Personalization KEY BUSINESS AND TECHNICAL REQUIREMENTS Collect demographic, contextual, and behavioral data to increase visitor profile accuracy Manage hundreds of millions of visitor profiles to improve visitor engagement Continuously improve the accuracy of visitor profiles by adding new attributes COUCHBASE SERVER NoSQL SOLUTION Document-level locking and high throughput, low latency writes supports massive number of concurrent visitors Flexible data model based on JSON documents enables agile development Managed, integrated object cache speeds read and write performance CUSTOMER HIGHLIGHTS AdAction replaced Cassandra with Couchbase Server for a 10x improvement in read latency PayPal maintains over 1 billion identities and 650 million profiles to create personalized experiences CUSTOMER SPOTLIGHT AOL Powers Billions of Targeted Ads Using NoSQL Advertising.com, owned by AOL, is one of the largest ad networks in the world with billions of impression per month and hundreds of millions of visitors. Clickstream data is fed to Hadoop for analysis, generating visitor profiles that are stored and accessed in real time from within Couchbase Server. Advertising.com leverages the Couchbase Server plugin for Sqoop, certified by Cloudera, to import and export data to and from Hadoop. In addition, after ads are served, visitor profiles are updated to improve ad targeting and placement. The integrated cache in Couchbase Server allows visitor profiles for active campaigns to be resident in memory to deliver sub-millisecond response times, while the flexible data model enables Advertising.com to continuously refine targeting algorithms by enriching the data. Personalization Personalization is fast becoming an opportunity, and a challenge. It’s an opportunity to make the right engagement – an ad, a coupon, a recommendation, and more – with the right visitor at the right time. But the ability to ingest, process, and utilize the amount of data necessary to create personalized experiences is a challenge for relational databases. Relational databases are overwhelmed by the volume of data required for personalization A personalized experience requires data, and lots of it: Demographic, contextual, behavioral, and more. The more data available, the more personalized the experience. A significant source of data is clickstream data – i.e., high volume, high velocity data – the type of write-intensive workload that overwhelms relational databases. In contrast, a distributed NoSQL database like Couchbase Server can elastically scale to meet the most demanding workloads. Rigid schemas of relational databases make it difficult to build and update profiles on the fly After ingesting visitor data, the next challenge is maintaining accurate visitor profiles. It’s not uncommon for visitor profiles to include hundreds if not thousands of attributes, as the number of attributes constantly grows. Increasing the number of attributes improves accuracy, but it can result in hundreds of millions of visitor profiles. While relational databases rely on fixed data models and offer limited scalability, document databases rely on flexible data models and provided unparalleled scalability. Relational databases are unable to deliver the low latency required for real-time responsiveness Finally, application logic must be able to access visitor profiles within milliseconds. The application must be able to engage visitors in real time. It’s not only a user expectation – it’s a business requirement. Advertisers, in particular, must meet service level agreements or else an ad will not result in an impression (and therefore no revenue) if it’s not delivered on time. And even more important than throughput is latency: While relational databases are limited to the speed of disk, a database like Couchbase Server, with integrated caching, can operate hundreds of times faster, at the speed of memory. Top 10 Enterprise NoSQL Use Cases | 7 Profile Management KEY BUSINESS AND TECHNICAL REQUIREMENTS Meet user experience expectations with high performance access to user profiles Scale to support continuous increase in both the number of users and the number of user interactions Leverage user profiles to improve engagement by adding new attributes on demand COUCHBASE SERVER NoSQL SOLUTION Managed, integrated object cache delivers high read/write performance Flexible data model based on JSON documents enables agile development Scale out architecture, with an integrated admin console, enables “push button” scaling on demand CUSTOMER HIGHLIGHTS Social Gaming Network scaled Couchbase Server to support the success of Cookie Jam, a title that exceeded 35 million users in just 10 months eBay maintains metadata for over 150 million users and over 100 million tokens for authenticated users in Couchbase Server for an improved user experience CUSTOMER SPOTLIGHT Fortune 100 Company Uses NoSQL to Manage Hundreds of Millions of User Profiles One of the world’s largest consumer electronics makers, a Fortune 100 company, manages several hundred million user profiles with a Couchbase Server deployment across multiple data centers globally. The company evaluated both MongoDB and Cassandra, and decided to replace Oracle with Couchbase Server as part of a multi-phase process to achieve better performance at scale and faster, less complex cross data center replication (XDCR). They began by deploying Couchbase Server as the primary user profile data store, with Oracle as a backup. In successive phases, they will leverage Couchbase Server to speed and simplify XDCR and to replace Oracle as a backup. Profile Management User profile management is core to web and mobile applications. It enables online transactions, user preferences, user authentication, and more. As the number of users, complexity of user profile data, and user experience expectations accelerate, relational database technology struggles to keep up with scalability, data flexibility, and performance requirements. NoSQL scale-out architecture enables faster, easier, more affordable scalability Today, web and mobile applications support thousands if not millions – or even hundreds of millions – of users. While relational databases serve user profile data with a single server, distributed databases deploy multiple servers: Capacity is increased simply by adding servers, making it far easier and less expensive to scale than with relational databases. NoSQL document databases provide flexibility to handle changing data types User profile complexity is also increasing with advances in technology, new features and services, and the desire to leverage more data – e.g., demographics, preferences, history, behavior, and more. While user profiles in relational databases are defined by rigid data models – which means complex and time- consuming schema changes when new attributes or data types are added – document databases support dynamic data models, enabling fast and easy changes on the fly. NoSQL with integrated cache delivers faster read/write performance than relational Finally, the availability of broadband Internet, the evolution of web browsers, and the rise of mobile apps have led users to expect an immediate response when they create a profile or login. While relational databases rely on separate caches to improve read performance – resulting in more expensive, complex, and harder to manage solutions – Couchbase Server features an easy-to-manage, integrated cache that delivers blazing fast read and write performance. Top 10 Enterprise NoSQL Use Cases | 8 Real-Time Big Data KEY BUSINESS AND TECHNICAL REQUIREMENTS Act on new data as soon as possible to improve operational efficiency Make actionable information available via dashboards and other real-time applications Maintain a single solution that meets both operational and analytical requirements COUCHBASE SERVER NoSQL SOLUTION Integrates with distributed messaging and stream processing systems like Kafka and Storm Provides proven, certified integration with Hadoop distributions from Cloudera and Hortonworks Memory-centric architecture delivers high speed reads and writes to meet constantly growing performance requirements CUSTOMER HIGHLIGHTS AOL integrates Couchbase Server with Hadoop to enable real- time ad targeting LivePerson processes terabytes of continuous clickstream data to monitor visitors in real-time CUSTOMER SPOTLIGHT PayPal Monitors Site Traffic in Real Time with NoSQL PayPal integrated Couchbase Server, Storm, and Hadoop to build the foundation of its real-time analytics platform. Clickstream and interaction data from all channels is pushed to the platform for real-time analysis. The data is pulled in a stream processor, Storm, for filtering, enrichment, and aggregation. After the data is processed, it’s written to Couchbase Server where it’s accessed by rich visualization tools. With these tools, PayPal can monitor all traffic in real-time. In addition, PayPal leverages views in Couchbase Server to perform additional aggregation and filtering. Finally, the data is exported from Couchbase Server to Hadoop for offline analysis. Real-Time Big Data The ability to extract information from operational data in real time is critical for an agile enterprise — in particular, the ability to increase operational efficiency, reduce costs, and increase revenue by acting on current data immediately. Hadoop is engineered for big data analytics, but it’s not real time. NoSQL is engineered for real-time big data, but it’s operational rather than analytical. Using NoSQL together with Hadoop is the answer for real time big data. Operational and analytical databases need to work together In the past, operational databases and analytical databases were maintained as different environments. The operational database powered applications. The analytical database was part of the business intelligence and reporting environment. They were not well integrated. They relied on a slow and complex process to load data from the operational database into the analytical database. The analytical database was never current. Modern big data infrastructure: NoSQL plus Hadoop Today, enterprises are moving from batch processing to stream processing. While batch processing is slow and incremental, stream processing is fast and continuous. A stream processor such as Storm analyzes data the moment it’s generated. However, it does not analyze historical data. That’s what Hadoop is for. Nor does it provide access to the results, which is necessary for enterprises to take action. That’s why a NoSQL database is necessary – as a front-end to store and manage operational data from any source and feed data to Hadoop, as well as a back-end to receive, store and serve analytic results from Hadoop. As a result of these requirements, modern big data infrastructure integrates a NoSQL database with Hadoop and a stream processor like Storm. Couchbase Server certified for leading Hadoop platforms – Cloudera and Hortonworks Couchbase Server, as a high performance NoSQL distributed database, is certified to integrate with leading Hadoop platforms including Cloudera and Hortonworks. Couchbase Server can be utilized with popular tools like Storm, as well as Sqoop and Kafka, to provide seamless Hadoop integration for a complete big data solution. Top 10 Enterprise NoSQL Use Cases | 9 Content Management Regardless of industry, every enterprise needs to engage users with rich, informative content. It’s a fundamental requirement of doing business today. Content isn’t limited to text or neatly structured data. It includes all kinds of semi- and unstructured data: Images, audio, video, presentations, and much more. And it’s not limited to content produced by the brand – it includes user-generated content such as photos, videos, reviews, ratings, and comments. Relational databases struggle to manage diverse content types due to their fixed data model The key to effective content is combining it – the ability to select a variety of content, aggregate it, and present it to the customer at the moment of interaction. But due to the fixed data model in relational databases, it’s difficult to create, maintain, and modify content like product pages, support portals, knowledge bases, and online magazines using relational technology. It’s not easy to add new types of content, because it requires a new data model, nor is it easy even to add new attributes, since it requires changes to the data model. Flexibility of document-oriented, NoSQL databases supports any type of content KEY BUSINESS AND TECHNICAL REQUIREMENTS Support the creation of new types of compelling content without administration overhead Engage customers by presenting them with a variety of content in multiple formats Meet user expectations, web or mobile, by providing immediate access to the latest content COUCHBASE SERVER NoSQL SOLUTION Flexible data model, based on JSON documents, stores any kind of content – structured, semi- and unstructured Integrated cache provides high throughput, low latency access to content Scale out architecture supports large volumes of content simply by adding servers on demand CUSTOMER SPOTLIGHT Fortune 500 Media Company leverages NoSQL to serve content to 50M online visitors A Fortune 500 media company with 90+ media outlets replaced Microsoft SQL Server with Couchbase Server to support new content types, semi- and unstructured, while improving the overall online experience for 50+ million unique monthly visitors. The company’s publishing tier, providing access to centralized web-facing data for all its properties, now delivers a sustained throughput of 50K reads and 10K writes per second, allowing them to easily keep up with demand. By contrast, a NoSQL document database, with its flexible data model, is perfect for storing rich, complex content. It can store and provide access to any type of content – structured, semi- or unstructured – because it does not require the data model to be defined first. This not only enables enterprises to easily create and produce new types of content, it also enables them to incorporate user-generated content – such as comments, images, or videos posted on social media – with the same ease and agility. Content management requires high performance and scalability, pushing the limits of relational databases It’s one thing to store content; it’s another to serve it. To be effective, applications have to deliver content to users immediately. For relational databases, this can be difficult as both the number of users and the amount of content increases. While relational databases are limited to how many users and how much content they can store on a single server, a distributed NoSQL database like Couchbase Server is not limited. Couchbase Server scales out simply by adding more servers. And with its integrated cache, Couchbase Server provides extremely high read performance by serving content at the speed of memory. Content Management Top 10 Enterprise NoSQL Use Cases | 10 Catalog Catalogs are not only referenced by web and mobile applications, they enable point-of-sale terminals, self-service kiosks, and more. In addition to product and service catalogs, there are reference catalogs for financial assets, employee data, digital media, and more. Catalogs may often contain user-generated content such as images, reviews, ratings, and comments. Rigid schema of relational databases makes data aggregation difficult As enterprises offer more products and services and collect more reference data, catalogs become fragmented by application and business unit or brand. Because relational databases rely on fixed data models, it’s not uncommon for multiple applications to access multiple databases – which introduces complexity and data management challenges. By contrast, a NoSQL document database, with its flexible data model, enables enterprises to more easily aggregate catalog data within a single database. NoSQL, document-oriented database enables on-the-fly data model changes Catalogs are also dynamic. Adding new products and services KEY BUSINESS AND TECHNICAL REQUIREMENTS Store catalog data for multiple applications and business units / brands within in a single database Add new products, services, and reference data or features that rely on them, on demand Ensure catalog data is available to applications during peak periods COUCHBASE SOLUTION Flexible data model based on JSON documents enables fast, agile development Managed, integrated object cache provides low-latency read performance Scale out architecture, with an integrated admin console, enables “push button” scaling on demand CUSTOMER HIGHLIGHTS Amadeus maintains the world’s largest database of travel- related data and processes more than 2 million requests / second with Couchbase Server Online travel provider Orbitz accesses hotel availability and rates from Couchbase Server CUSTOMER SPOTLIGHT NoSQL powers product catalogs and more for mega-retailer Tesco Tesco, the world’s largest Europe-based retailer, is adopting Couchbase Server for multiple use cases, including a centralized service for storing and managing data for millions of products, which can be leveraged across the company. Other targeted use cases for Couchbase Server include NoSQL databases to power pricing and inventory management applications, as well as an application to deliver customer-specific coupons and vouchers at the point of sale and on the website, and a customer orders database to track purchase history across all channels. Tesco is migrating to Couchbase’s distributed NoSQL platform in order to provide the flexibility to accommodate a diverse range of fast-changing data, as well as the performance, availability, and scalability to support massive numbers of users, large amounts of data, and very high throughput requirements. As Tesco’s requirements grow, Couchbase Server provides easy, push button scalability, allowing the retailer to simply add nodes on demand as system capacity and throughput needs increase. may require changes to the data model and new attributes, if not entirely new data models. The same is true for new features such as recommendations, promotions, and more. While relational databases are limited to static data models – which are complex and time-consuming to change – a document database enables on- the-fly updates to the data model, and thus dynamic catalogs. Relational databases are challenged to deliver high-performance read access Applications require high-performance read access to catalogs from multiple channels – point-of-sales, kiosk, web, mobile, and more – to support millions of users, which is a challenge for relational databases. A delay in accessing product information could cause a line to form at a checkout terminal or visitors to abandon a website or shopping cart. Couchbase Server addresses those requirements with its memory-centric architecture and automatic replication to ensure high performance and availability. A case in point: While many retailers struggled to maintain availability during Black Friday and Cyber Monday in 2014, one of the world’s largest mega-retailers who deployed Couchbase Server didn’t experience those problems and set record online sales. Catalog Top 10 Enterprise NoSQL Use Cases | 11 Customer 3600 View Creating and maintaining a unified view of the customer has been an important but elusive goal for enterprises. Customers expect a consistent experience, regardless of channel. And the enterprise wants to capitalize on upsell / cross-sell opportunities and provide the highest level of customer service in a cost effective manner. However, as the number of products and services, channels, brands, and business units increases, customer data becomes fragmented in silos. Use of relational databases leads to fragmented customer data The fixed data model of relational databases forces enterprises to fragment customer data, because different applications work with different customer data. As a result, these applications end up maintaining their own data model and a separate view of the customer. In contrast, a NoSQL document database utilizes a flexible data model that enables multiple applications to not only access the same customer data, but to add new attributes without affecting other applications. Relational databases are limited by their scale-up architecture As the number of customers grows from thousands to millions – and KEY BUSINESS AND TECHNICAL REQUIREMENTS Ensure a high quality experience for every customer via any channel - online, on the phone, or in-store Improve operational efficiency by maintaining a complete, consistent view of customer data Meet user expectations, web or mobile, for millions of customers COUCHBASE SERVER NoSQL SOLUTION Flexible, JSON-based data model supports the needs of multiple applications Integrated cache enables high throughput, low latency access to customer data Scale out architecture allows storage of large volumes of customer data by adding servers on demand CUSTOMER SPOTLIGHT Fortune 200 Global Apparel Brand Unifies Customer Data with NoSQL A Fortune 200 global apparel brand growing its direct-to-consumer online business faced scalability and flexibility challenges while trying to create and maintain a unified view of its customers. Each customer facing application had its own customer database, and the workload and number of online interactions was dramatically increasing. The company is deploying Couchbase Server to provide a centralized service consolidating customer data, which can be accessed and leveraged by multiple customer-facing and internal applications. Couchbase Server was chosen because it was simple to scale, met high performance and availability requirements, and supported the company’s shift to a public cloud for a distributed, elastic infrastructure. even hundreds of millions or more – enterprises need a scalable database to address not only a high volume of customer data, but a high number of customer interactions. While relational databases are limited to the resources of a single server (a “scale up” architecture), a distributed database can scale simply by increasing the number of servers (a “scale out” architecture). So as the number of customers grows, enterprises can just add servers on an as-needed basis. Relational databases struggle to deliver sub-millisecond response times When it comes to customer interactions, performance matters. Whether trying to make a purchase, view a billing statement online, or call for support, customers expect an immediate response. Unlike relational databases that rely on third-party caching solutions, Couchbase Server, with its integrated cache, can natively provide sub-millisecond response times required to deliver a great customer experience. Customer 3600 View Top 10 Enterprise NoSQL Use Cases | 12 Mobile Applications KEY BUSINESS AND TECHNICAL REQUIREMENTS Support viral growth (e.g., tens of millions of users within days or months) in a cost effective manner Launch new mobile applications with a faster time to market Provide users with a smooth, seamless experience, online or offline COUCHBASE SERVER NoSQL SOLUTION Scale out architecture, with an integrated admin console, enables “push button” scaling on demand Flexible data model based on JSON documents enables agile development Couchbase Mobile enables offline data access with automatic synchronization to deliver apps that always work with or without a network connection CUSTOMER HIGHLIGHTS CookieJam by SGN and Facebook Game of the Year scaled to 35 million users in ten months Draw Something by Zygna, scaled from 6 to 90 nodes in 8 weeks to support 15 million daily users Slotomania by Playtika supports up to 200K concurrent users during peak hours CUSTOMER SPOTLIGHT Ryanair re-platforms with Couchbase Mobile to improve user experience Ryanair, one of Europe’s largest and most successful airlines, is re-platforming its mobile app to support more than 1 million travelers, moving from a relational database to NoSQL using Couchbase Mobile. The company wanted to overcome the limitations of relational technology to significantly improve the quality, availability, and functionality of its mobile app and provide a great experience for mobile users. Couchbase Mobile includes Couchbase Server as the remote, backend database server; Couchbase Lite, a JSON database that’s embedded with the app on the user’s device; and Couchbase Sync Gateway, which enables two-way synchronization between Couchbase Server and Lite. Thanks to the embedded database, the solution enables Ryanair’s mobile app to work all the time – whether or not the user has a network connection – while Sync Gateway leverages network connectivity, whenever available, to keep the local and remote databases in sync. Mobile Applications Mobile phones and tablets are rapidly becoming the dominant platform for search, shopping, and other applications. Mobile users spend only 20% of their time in browsers and 80% within apps. Mobile applications present challenges with scalability, performance, and availability that relational database are ill equipped to address. Scale-out architecture of NoSQL enables elastic, affordable scalability With nearly 2 billion smartphone users, mobile applications face scalability challenges in terms of growth and volume. For instance, it’s not uncommon for mobile games to reach tens of millions of users in a matter of months. With a distributed, scale-out database, mobile applications can start with a small deployment and expand as the user base grows, rather than deploying an expensive, large relational database server from the beginning. NoSQL data model flexibility allows faster time to develop and update mobile apps A flexible data model is critical for mobile applications. In a market where applications go viral, time to market is essential. By adopting a flexible data model, mobile apps can be developed faster and launched sooner. In addition, a fixed data model can make it difficult to support multiple versions of a mobile application because different versions may have different data models. A document database solves this problem by enabling developers to create and maintain a single, flexible data model that can change and evolve on the fly, without the need to redefine schema beforehand. Mobile apps need to work with or without a network connection Users expect mobile applications to be highly interactive and highly responsive. This creates performance and availability challenges. The database must respond within milliseconds regardless of the number of users. However, a remote database in the cloud requires a network connection. For a mobile application to be available without a connection, it needs a local database and automatic synchronization with the remote database. Couchbase Mobile solves this challenge by extending Couchbase Server to mobile phones and tablets to support offline access. Top 10 Enterprise NoSQL Use Cases | 13 Internet of Things KEY BUSINESS AND TECHNICAL REQUIREMENTS Leverage IoT data to create new products and services Improve operational efficiency by collecting data from machines, equipment, and more Interact with connected devices in multiple geographic locations COUCHBASE SERVER NoSQL SOLUTION Document-level locking and memory-first writes support massive number of concurrent devices Flexible data model based on JSON documents supports new types of IoT data Built-in, streaming cross data center replication brings database closer to connected devices and end users Couchbase Mobile removes dependency on network access to continuously collect and store data CUSTOMER SPOTLIGHT Fortune 100 Telecom Delivers IoT App with NoSQL to Track Connected Devices A Fortune 100 telecommunications company chose Couchbase Server to power a new service enabling enterprise customers to monitor and manage devices connected to corporate networks. With the new service, customers can track and manage any type of connected device. In addition to phones, computers, and printers, the service can monitor other connected devices - alarms, lights, HVAC systems, locks, industrial equipment, and more - to provide customers with the visibility necessary to optimize device and network utilization for increased efficiency and reduced costs. The memory-centric architecture and integrated cache enable Couchbase Server to ingest usage data at high speed while stream based indexing enables it to create real-time views of the data. Internet of Things Today, some 20 billion devices are connected to the Internet – everything from smartphones and tablets to home appliances and systems installed in cars, hospitals, and warehouses. These devices receive data on environment, location, movement, temperature, weather, and more from 50 billion sensors. This data can be leveraged by enterprises in manufacturing, health care, retail, security, transportation, and many other industries. IoT applications require real-time access to operational data Innovative enterprises are leveraging the Internet of Things to develop new products and services, reduce costs and time to market, increase efficiency, eliminate waste, and boost customer satisfaction. This ability to access global, operational data in real- time enables dynamic, informed decision-making and increases business agility. Relational databases struggle with the volume, velocity, and variety of IoT data The volume, velocity, and variety of machine-generated data – a major contributor to the growth of big data and the Internet of Things – are increasing with the proliferation of digital telemetry, which is semi-structured and continuous. Relational databases struggle with the three well-known challenges from big data IoT applications: scalability (due to massive volumes of data), throughput (due to super fast data velocity), and data variety (structured and semi-structured). NoSQL technology is a better solution for IoT applications To overcome these challenges, innovative enterprises are relying on NoSQL technology like Couchbase Server to scale concurrent data access to millions of connected devices and systems, store large volumes of data, and meet the performance requirements of mission-critical infrastructure and operations. Couchbase Server provides the scalability to manage billions of data points, the speed and throughput necessary to ingest continuous streams of data in real-time, and the flexibility to store data that’s evolving every day. Top 10 Enterprise NoSQL Use Cases | 14 Digital Communication Today’s world relies on digital communication in the form mobile text messaging, online chat, real-time collaboration, and more. However, enabling millions of users to communicate in real-time requires both performance and availability at scale, and it’s a challenge for relational databases. Digital communication apps require massive scalability – a weak point for relational databases Mobile text messaging applications are hugely popular. They have to support millions if not hundreds of millions of users. That means hundreds of thousands to millions of requests per second, and users expect immediate responses. It’s not acceptable to wait seconds for a message to be sent. It must be sent within milliseconds. This type of write-intensive workload with many concurrent users is a challenge for relational databases. While the scalability of relational databases is limited to increasing the amount of memory or processors within a single server, a distributed NoSQL database like Couchbase Server easily and elastically scales as more servers are added. Relational databases cannot deliver real-time responsiveness required for digital communications KEY BUSINESS AND TECHNICAL REQUIREMENTS Enable millions of users to send and receive messages in real time Ensure digital communication platforms are available 24x365 Support a global user base with users in multiple regions and countries COUCHBASE SERVER NoSQL SOLUTION Flexible data model based on JSON documents enables agile development Managed, integrated object cache delivers high read/write performance Scale out architecture, with an integrated admin console, enables “push button” scaling on demand CUSTOMER HIGHLIGHTS Viber replaced MongoDB with Couchbase Server to handle billions of messages per month LivePerson replaced Cassandra with Couchbase Server to manage 20M+ engagements per month CUSTOMER SPOTLIGHT LivePerson Leverages NoSQL to Power Real-Time Chat LivePerson helps over 8,500 enterprise customers better engage their website visitors. They considered MongoDB and Cassandra, but chose Couchbase Server because it met all of their requirements including high availability, full text search, document handling, key/value storage, high performance, cross data center replication, and more. Their agents rely on a dashboard to monitor and chat with website visitors in real time. The dashboard is powered by Couchbase Server views to sort, filter, and aggregate streams of website visitor data. In an enterprise environment, digital communication may take the form of online interaction via direct messaging to help visitors find a product or complete the checkout process. This type of online interaction must take place in real time. If it’s delayed, visitors may abandon the website before a representative can interact with them. And as with mobile text messaging, the application may need to support millions of website visitors. While the responsiveness of relational databases is limited by operating at the speed of disk, Couchbase Server, with its integrated cache and memory-centric architecture delivers the sub-millisecond responsiveness that digital communications applications require. Meeting the need for always-on availability In a connected world, users are no longer confined to a specific region or country. In addition to being available 24 hours day, 365 days a year, digital communication platforms must be available to users in multiple regions and countries. While relational databases require third-party software for cross data center replication (XDCR) – resulting in a slow and complex solution – Couchbase Server features integrated, streaming XDCR for high availability, disaster recovery, and data locality. Digital Communication Top 10 Enterprise NoSQL Use Cases | 15 Fraud Detection For financial services enterprises, fraud detection and risk analytics are essential to reducing profit loss, minimizing financial exposure, and complying with regulations. The process impacts both the enterprise and its customers: It relies on data – rules, customer information, transaction information, location, time of day, and more – applied at scale and in real-time. Sub-millisecond responsiveness required for real time fraud detection When customers pay with a credit or debit card online or in- person, they expect immediate confirmation. However, the transaction must be processed by a fraud detection platform, which has to access customer data and fraud detection rules in less than a millisecond. It’s a low latency requirement that relational databases fail to meet. In contrast, Couchbase Server is able to provide data access at the speed of memory by leveraging an integrated cache. Massive scalability requirements The financial services industry serves hundreds of millions of KEY BUSINESS AND TECHNICAL REQUIREMENTS Apply fraud detections rules to financial transactions in real- time Scale to support millions of card holders and billions of transactions Ensure fraud detection platforms are running 24x365 COUCHBASE SERVER NoSQL SOLUTION Managed, integrated cache enables sub-millisecond data access Scales to support billions of transactions simply by increasing the number of nodes Built-in, streaming cross data center replication delivers high availability and disaster recovery support CUSTOMER SPOTLIGHT NoSQL Powers a Leading Fraud Detection Platform A leading fraud detection platform, which processes more than 50% of the world’s credit and debit card transactions, is powered by Couchbase Server. Its customers include some of the largest financial institutions in the world, each with hundreds of millions of cardholders. The company selected Couchbase Server as a distributed database for the fraud detection platform, because its integrated, in-memory cache enables sub-millisecond access to customer data and fraud detection rules, and because it can scale with an increasing number of cardholders and transactions in a cost effective manner. customers who perform billions of transactions per year, in real time. While the ability for relational databases to operate at this scale is expensive and limited, distributed databases rely on commodity hardware with no limits on operating scale. Always on availability is a must As a mission critical application, the availability and disaster recovery requirements for fraud detection are exceptionally high. Applications must run 24 hours a day, 365 days a year, and require availability across data centers. While relational databases rely on complex, slow software to replicate data to multiple data centers, Couchbase Server features integrated, streaming cross data center replication to increase availability and support disaster recovery. Fraud Detection Couchbase delivers the world’s highest performing NoSQL distributed database platform. Developers around the world use the Couchbase platform to build enterprise web, mobile, and IoT applications that support massive data volumes in real time. The Couchbase platform includes Couchbase Server, Couchbase Lite — the first mobile NoSQL database, and Couchbase Sync Gateway. Couchbase is designed for global deployments, with configurable cross data center replication to increase data local- ity and availability. All Couchbase products are open source projects. Couchbase customers include industry leaders like AOL, AT&T, Bally’s, Beats Music, BSkyB, Cisco, Comcast, Concur, Disney, eBay, KDDI, Nordstorm, Neiman Marcus, Orbitz, PayPal, Rakuten / Viber, Tencent, Verizon, Wells Fargo, Willis Group, as well as hundreds of other household names. Couchbase investors include Accel Partners, Adams Street Partners, Ignition Partners, Mayfield Fund, North Bridge Venture Partners, and West Summit. About Couchbase 2440 West El Camino Real | Ste 600 Mountain View, California 94040 1-650-417-7500 www.couchbase.com Conclusion Four Key Takeaways This list of the top 10 NoSQL use cases, along with real-world examples, illustrates the extent to which NoSQL has penetrated the enterprise. NoSQL has clearly caught the interest and imagination of both developers and business managers as a solution to key challenges that relational database technology simply was never designed to address. There are four key conclusions to take away from these use case examples: 1. NoSQL has advanced well beyond experimentation NoSQL adoption has clearly moved beyond experimenters and innovators to mainstream “early adopters” for mission critical applications. While it’s still early days in the NoSQL adoption cycle, enterprises are using NoSQL for mission critical applications that drive revenue and serve massive numbers of users. 2. NoSQL is being adopted in virtually every industry NoSQL adopters referenced in this document represent a very broad range of industries, from retail and financial services to travel, telecommunications, and government. Enterprises that need to develop and deploy large-scale enterprise web, mobile, and IoT applications – which includes pretty much every industry – are turning to NoSQL as a key part of the solution. 3. NoSQL adopters are realizing significant, measurable benefits Enterprises are reporting major, quantifiable benefits from their use of NoSQL. Most importantly, they’re solving problems that traditional relational databases cannot address. Among other benefits, NoSQL is enabling enterprises to deliver more responsive applications; scale their applications more affordably; increase resource efficiency; develop applications more quickly; and harness the value of big data to better understand customers and improve their business. 4. NoSQL is a fundamental cornerstone of modern, big data infrastructure Along with other key big data technologies such as Hadoop, NoSQL has become a fundamental cornerstone of modern, big data infrastructure. Already we are seeing a number of forward-looking enterprises – those known for their early use of emerging technologies – begin to “re-platform” their technology infrastructure with NoSQL to build a new foundation that meets big data requirements. Over the next several years, we expect to see this broad re-platforming take off in a big way.