X

Nerve.com, an online dating and content site, wanted to build a new type of dating app that used social content and conversations to mimic dating interactions in the physical world. They planned to use the latest web app technologies, using a Nodestack deployment of Node.js application servers and MongoDB databases. However, they wanted to remain focused on building apps and not waste time maintaining and administering databases, application servers and hardware. They didn't have any dedicated systems administration specialists nor did they want any. Data reliability and security were paramount for the site. The real-time nature of interactions on social.nerve.com required instantaneous response times from both the database and the application server to keep users happy. To maintain both performance and reliability would require excellent support and uptime.

Message Bus, a cloud-native application company, was seeking to provide a service that could eliminate the challenges around building email or mobile messaging capabilities into any software or web application. They wanted to offer a viable alternative to using out-of-the-box enterprise messaging software. However, to continue telling a compelling cloud story in the marketplace, they needed to work with a cloud infrastructure provider that could truly deliver all the advantages of the cloud. They needed scalability, consistent performance measurement across their network, CPUs, and disks, support for new standards and technologies, and a true business partner.

Reebonz, an exclusive online shopping destination, was facing repeated problems when the response to popular sales crashed their Web infrastructure. The company releases deals at noon Singapore time, which resulted in a tidal wave of pageview requests in a very concentrated time. This led to many outages and loss of revenues during the peak hour. The company needed a solution that could respond to demand spikes by scaling quickly and massively. They also wanted a cost-effective cloud computing solution and strong migration support for moving their significant infrastructure.

ModCloth, an online retailer specializing in vintage-inspired women’s fashion and decor, was experiencing rapid growth and needed a highly scalable, agile, and consistent cloud infrastructure to support its ecommerce website. The company's site experience is built around a quick and agile inventory strategy that involves launching 25 to 50 items every day, resulting in 30% of customers visiting more than once per day. Therefore, the performance of the website was a business imperative. ModCloth needed visibility into its infrastructure to solve performance issues and ensure the applications were built correctly for future performance on the platform. The company also needed to deliver a consistently great experience to customers, regardless of how busy the site was or how congested the network might become. Furthermore, ModCloth needed the ability to make changes quickly to take advantage of new opportunities and respond to customers’ changing behaviors.

Quizlet, a rapidly growing online study platform, was seeking a cloud platform to build its business on. The company required a solution that offered data reliability and 100% uptime, superior analytics tools for faster page loads, premium hardware in the data center, and responsive support. The company's users would be accessing the site from various devices and locations, exposing them to a diverse set of data delivery and network conditions. Therefore, Quizlet needed to quickly analyze and troubleshoot any application latencies to tune apps for high-speed page delivery. The company also wanted to ensure proactive fixes for any problem to prevent cascading failures that affect end users.

Moonshadow Mobile needed an infrastructure that could allow it to deliver highly detailed geospatial data in online mapping applications in near real time for a highly demanding customer base paying big bucks for unique applications. The company offers over 200 real-time web and mobile applications for customers and delivers population data in numerous ways and forms. The challenge is that usage is variable. So if you don’t have to over-provision servers for massive spikes in usage but still maintain performance, you will save a lot of money. Moonshadow also wanted high-speed internal networks to push huge volumes of data between the database and the application networks as well as a way to easily distribute its workloads across multiple application servers to ensure customers always got fast response times via the Internet.

Nodejitsu, a premier Node.js platform as a service provider, was looking for a cloud infrastructure provider that could offer deep knowledge of Node.js, easy account management, better performance, and high reliability. They wanted to offer the highest levels of performance to its clients and needed a cloud provider that could support their infrastructure-agnostic service that automatically scales to support the needs of applications. They also wanted to avoid the hassle of managing multiple hosting accounts once they exceed a certain number of servers.

Voxer, a social networking application that turns your phone into a walkie-talkie and an all-in-one messenger, faced a challenge when its user base grew 10X over the course of a month. Their Linux-based storage system could not handle the load, leading to peak times where users had to wait for their messages to be downloaded, instead of always streaming them live. Voxer's back-end software is written entirely in Node.js, which can be challenging to understand and operate in production and at scale. The company needed to improve performance and gain complete transparency into all critical processes in the Voxer architecture and application stack, from low-level CPU and disk processes to database queries to web application servers and HTTP operations.

Cargill, a global leader in the food, agriculture, financial and industrial sectors, was facing challenges in creating new digital products and services due to its legacy IT systems. The company's existing systems were slowing down its ability to innovate and meet the evolving needs of its customers and partners. Cargill recognized the need for a more dynamic and accessible internal suite of APIs to enable consumption across the entire company. The company wanted to transition from siloed legacy IT architecture to a modern, microservices architecture. However, the legacy API management vendors they considered were not aligned with their vision for decentralization.

Digicel, a telecommunications and home entertainment provider, embarked on an effort to launch a fully digital mobile network for its French West Indies region. The company's existing technology stack relied on a VPN network and ESB to expose services and faced issues with connectivity and functionality in microservices use cases. The company needed to adopt a microservices architecture to support this transformation and remain competitive. They were looking for an API gateway that would support their domain-driven design approach and work well with their architecture.

SoulCycle, a leading fitness market leader with more than 50,000 riders a week attending classes, was facing challenges due to its monolithic architecture which was a critical bottleneck to service development and scalability. With aggressive growth targets, SoulCycle needed to leverage technology to better integrate with partners and enable expansion into new business models. To address these challenges, Jason Rodriguez, VP of Engineering, and his team created the SoulCycle Podium Platform to support agile development for all internal and externally-facing applications. As part of this transition, Rodriguez and his team quickly realized the need for an API platform that could efficiently support SoulCycle’s more than 30 services, utilities, and monoliths across multiple clouds at scale.

Yahoo! Japan Corporation offers more than 100 services, including search engine, auction, news, weather, sport, email and shopping. Yahoo! Japan is the undisputed market leader in terms of page views with more than 70 billion each month. To maintain market leadership, Yahoo! Japan began transitioning towards microservices to accelerate service development and deployment. Consequently, Yahoo! Japan needed a new API platform to centralize its portfolio of APIs and to improve the security, availability and performance of its services. “We moved rapidly towards microservices over the past few years, and we needed to centralize our APIs,” said Kanaderu Fukuda, senior manager of the Computing Platform Department at Yahoo! Japan. “We began looking into API gateway products, and after comparing several products, decided to use Kong.”

Cort, a global leader in furniture rental and related services, standardized on the use of Kubernetes for important workloads such as Magento for e-commerce and BitBucket for their code repositories as well as MariaDB, PostgreSQL, and other databases. They wanted to fully leverage and preserve benefits such as separation of concerns between applications and infrastructure, quick and consistent instantiation of environments, and enhanced operational efficiency benefits typically gained from using Kubernetes for compute. However, they found that despite extremely vocal support of ‘Cloud Native’ and Kubernetes, e.g., Container Storage Interface (CSI) adoption, all considered vendors were actually not Kubernetes native. They operated and behaved entirely differently than Kubernetes itself, increasing the complexity of the environment while creating a shared dependency for all stateful workloads. Additionally, to simplify operations and improve the resilience of their environments, including backing up and enabling disaster recovery across two domestic data centers, Cort wanted to have an easy to use application to centralize their management of Kubernetes as a data infrastructure management layer.

The insurance company was looking to improve their development agility by accelerating their adoption of containers as a service for their internal teams. They had been using Docker Swarm for stateless workloads, but with the emergence of Kubernetes and Kubernetes services such as Azure Kubernetes Services, they felt it was appropriate to begin to run stateful workloads on containers. However, they faced limitations in the performance and flexibility of underlying Azure storage, and Azure managed disks. They also encountered technical and performance limitations of Azure file, which is limited with SMB protocol only and can be tricky to manage.

Skullcandy, a leading audio brand, launched its eCommerce site in 2008. The site was built on the Magento eCommerce platform with PHP and Linux. The rapid growth of online sales put pressure on the interactive team to keep up with the increasing demands. The team's goal was to ensure that their online customers have a quick and easy purchase transaction every time they visit the site. However, the site was experiencing growing pains with a number of performance issues. Unplanned outages were on the rise, usually occurring more than once a week and often for significant duration. These outages equated to lost revenue, and potentially, even to lost customers. The team was still learning about performance issues and problems from their customers. Without any proactive real-time performance monitoring and management tools, they simply were not able to see what was going on in real time.

BlackLine Systems, a provider of enterprise-class software used to automate and control the financial close process, was facing challenges with its growing business and expanding global customer base. The company needed reliable high-speed performance to ensure delivery of its SaaS-based month-end close and reconciliation application to clients. Serving companies of all sizes that must adhere to compliance-related financial deadlines, BlackLine Systems had self-imposed performance and uptime requirements designed to meet customers’ varying needs. However, ensuring superior performance became more challenging as BlackLine’s customer base continued to grow rapidly. For the BlackLine app team, understanding application performance and having the ability to quickly find and diagnose performance issues before they affect customers became increasingly important.

Zumba Fitness, a global lifestyle brand that fuses fitness, entertainment, and culture into an exhilarating dance-party workout, has been experiencing rapid global growth with more than 50% of its 2011 growth coming from outside the United States. Zumba’s online stores are available on every continent and in more than 80 countries around the globe. With self-imposed performance objectives, the Zumba team is keenly aware that superior performance is essential for ensuring exceptional online service for both its B2C and B2B customers. Rapid growth has presented app performance challenges to the Zumba team, particularly in getting a clear picture of end user performance in various locations around the globe. The team had been using a combination of homegrown and open source tools, which required working across multiple environments – a cumbersome and time-consuming process. Additionally, mobile access created its own performance challenges. Zumba was finding it increasingly difficult to stay ahead of performance issues, let alone preventing them.

Uken, a cross-platform game developer, has eight games in production that have been downloaded more than 20 million times. The games are browser-based applications that push the limits of browser capabilities. The company processes about 800 Million requests each month. All common processes needed to run its games and business, such as user authentication, notifications, payments, analytics, and an ad platform, are pulled out and made into one of the twelve internally shared services. With self-imposed performance goals, Uken used basic tools and a few small Rails plugins to monitor and manage performance before they installed New Relic. This approach meant the team was spending valuable time performing manual work to determine what was going on and why some users were experiencing slow response times. The company quickly recognized that it needed some additional help.

Paper Culture, an eco-friendly stationery company, was facing challenges in maintaining 100% uptime and fast performance for its customers and search engines. The company often experienced seasonal spikes in traffic due to promotional email blasts, which could go viral and create a lot of additional traffic. This, coupled with customers uploading a wide range of complex image files, resulted in peak processing loads. As their business grew, Paper Culture was motivated to continually improve performance. The team had adopted an agile development strategy, often deploying daily enhancements, frequently on demand. It was critical for the technical team to understand the impact of every change on site performance. In the past, Paper Culture often learned about slow performance or other issues from their customers. When Paper Culture received one of these reports, they had to dive into logs for hours, going through thousands of lines of code and piecing a chain of events together to understand and resolve the issue. The technical team knew they needed better insight and control over the development and deployment environments if they wanted to achieve their goal of 100% customer satisfaction.

Nike's online business is a significant and growing part of its revenue portfolio. In 2011, Nike.com and its associated eCommerce applications brought in about $450 Million in revenue in the United States and about $200 Million plus in Europe. However, prior to the company’s bringing in New Relic in 2011, incident notification, performance triage and management were significant challenges for the nike.com eCommerce team. There was very little insight into production application server health. And although the team was using tools, they didn’t provide the detailed insight the team needed to ensure performance of Nike’s rapidly growing eCommerce platform. As Kevin Bartholomew, Nike Web Support Production Manager says, “Our main challenge was insufficient insight into our eCommerce applications — no single pane of glass we could look at to see what was going on under the covers. One tool told us the network was down. Another told us the site was slow. We didn’t have anything to see where the problems were, and definitely not what was causing potential issues.”

Cinchcast, a company that provides a cloud-based audio management platform, was experiencing rapid growth as more individuals, businesses, and organizations discovered how easy it was to create and deliver audio content using their solution. On a monthly basis, Cinchcast’s media property, BlogTalkRadio, was hosting more than 35,000 hours of original content, attracting over 13 million unique visitors, and powering 15 million streams and 175 million ad impressions. The team at Cinchcast knew that site performance was critical to participant engagement and saw it as a critical success factor for their continued rapid growth. They needed more insight and visibility into application performance to understand which components in the application were taking significantly more time than others and how fast the website loaded from an end-user perspective. Using multiple tools from multiple vendors was time-consuming and did not offer a seamless monitoring experience.

Bleacher Report, a digital sports media outlet, has over 26 million readers who rely on the site for news about their favorite sports and teams. During peak times like the Super Bowl, World Series, March Madness, etc., fans flood the site for up-to-the-minute information and dialogue. High performing mobile access is especially critical. March is a particularly busy month in terms of traffic to the site: NCAA March Madness is right on top of the NFL free agency and the NBA trading deadline. During March Madness, the site processes 50,000 requests per minute (RPM) for the core service. With a self-imposed uptime goal of 99.9% and striving for 99.99% uptime, the B/R engineering team needed a real-time, all-the-time, application monitoring tool.

Fanzter, a private software company, was under pressure to maintain high performance for its products, introduce new features, and increase application stability. The company's products, which include Coolspotters.com, CoolPapers, Streaks, and Summizer, served over 20 million users in 2011, serving more than 32 million page views each month. High performance for traditional and mobile access was essential to their success. In the early days of Coolspotters, the team analyzed Rails application logs daily to determine the areas with slowest performance. They also tried to correlate the slow query logs from the database with the actions that triggered them. These methods were time consuming and far from foolproof. Fanzter needed a better way to analyze performance, to find and fix performance problems.

ProSoft Technology, a company specializing in communication solutions for industrial automation and control applications, was looking to develop a new cloud-based communication platform for the Industrial Internet of Things (IIoT) market. The company wanted to provide its customers with a secure, flexible, and easy-to-use platform that would allow them to remotely monitor and manage their industrial systems. However, ProSoft faced several challenges in developing this platform. They needed a solution that would allow them to quickly and easily develop and deliver new protocols, processes, and technologies that meet the needs of their industrial customers. They also needed a solution that would provide them with the flexibility to add additional services as their customers' needs evolve.

TrustBills, a German-based auction platform for trade receivables, faced several challenges in its operations. The company had to comply with stringent EU data privacy and compliance regulations, which required them to run their own Kubernetes cluster instead of using a managed service. This was further complicated by the need to protect and secure customer data in a security-conscious market. TrustBills spent 10 months testing open source container storage solutions but failed to identify a solution with sufficient levels of scalability, resilience, and security. The company needed a solution that would allow them to efficiently allocate resources, provide performance guarantees for each service, and ensure that if one service was compromised, not all other services would be affected.

Beco, a company that delivers real-time space analytics to connect workforces to physical spaces, faced challenges in processing large amounts of data collected from connected devices. They needed to keep important databases like Kafka, Cassandra, and PostgreSQL highly available when using containers. The default setting of DC/OS pins stateful services to a single host, which posed a risk to high availability. The company needed a solution that could handle stateful services in a reliable, scalable fashion.

Cloud Provider, a cloud hosting, infrastructure, and services company, was facing challenges in offering scalable solutions to its customers. The company's customers, who are mostly not tech-savvy, were finding it difficult to migrate from a shared hosting environment to a virtual machine. The company wanted to offer a solution that would allow its customers to scale their operations without having to manage anything. The main challenge was ensuring the scalability of the platform, especially for stateful apps like WordPress. For these apps to function properly, it was necessary that the data volume is available to multiple physical hosts at the same time. This was a requirement that many storage vendors did not offer.

Fractal Industries, a software company that applies artificial intelligence to solve complex, real-world problems at scale, faced a significant challenge in managing its stateful services. The company's platform, Fractal OS, needed to run in multiple environments, including their own multi-tenant SaaS infrastructure, a customer's data center, or their VPC in Amazon, Google, or Azure. However, most container orchestration platforms are built for stateless services, which are easy to scale elastically. Existing persistent storage and data management solutions didn't work across clouds and on-premises data centers, a hard requirement for Fractal's platform. Furthermore, running multiple data services at production scale required significant expertise in each service to provide high availability, backups, and disaster recovery.

MightWeb, a hosting company, needed to instantly scale to meet customer demand, which was challenging for the many stateful services they offer like WordPress, MySQL, MongoDB, Magento, and WooCommerce. Most persistent storage options for containers don’t provide both high performance storage for applications like MySQL, and multi-writer shared volumes for WordPress, both hard requirements for typical hosting customers. They needed a solution that could provide high availability, the ability to take snapshots, and the ability to scale.

Lix, a study platform for university students, was looking for a way to improve their platform's responsiveness and speed. They wanted to use containers to build a high-quality platform, but they faced challenges in running and managing data services like Elasticsearch, Cassandra, and Postgres in containers. Initially, they manually managed node labels for these services, which was time-consuming and prone to errors. Moreover, when a machine failed, they lacked the ability to temporarily move the Cassandra pod to another machine while the cluster recovered.