Last Friday we hosted ThousandEyes Connect New York at a restored 1902 space overlooking Bryant Park. We welcomed attendees from top US and global banks, insurance companies, media companies, cloud infrastructure providers and healthcare companies. What all of them had in common was in looking beyond traditional network monitoring in order to tame environments that have become too complex, mission critical and hidden. The event included engaging talks from Credit Suisse, DigitalOcean and Shutterstock. More on each of the talks in a bit, but first let’s set some context.
A New Community Solving Complex Network Problems
Last May, we hosted ThousandEyes Connect San Francisco. The goal was simple yet bold: to bring together the brightest networking minds solving the toughest networking problems. We invited some amazing companies to talk.
- Bloomberg talked about delivering "better than best effort reliability on the Internet."
- eBay talked about "optimizing CDN and web performance."
- Oracle talked about managing their "public cloud operations."
- Twitter spoke about "keeping the tweets flowing."
Follow the links to read the deep dive blogs! Needless to say, feedback from the event was so positive that running another one became an inevitability.
Beyond Network Monitoring
Back to ThousandEyes Connect New York. For this event, we invited three amazing speakers from Shutterstock, DigitalOcean and Credit Suisse.
First up though, was Mohit Lad, our CEO and Co-Founder, who cut to the chase: “we've built ThousandEyes for chaotic networks you no longer control.” Meaning that we’re in a world today where networks span corporate, Internet and service provider segments, where large portions of it are hidden and dynamic. He explained how companies at the forefront are using ThousandEyes to accurately understand network behavior across all segments; to respond quickly when there’s an issue and before it impacts applications, services and customers.
Shutterstock
Our first customer speaker was Gene Yaacobi, Infrastructure Manager at Shutterstock. Shutterstock is the stock photo, footage and music company we’ve all heard of. Now with offices globally, and with 1.3 million active customers from 150 countries, they have thousands of images and videos being added every hour.
Impressive stats, but what does their infrastructure look like? Gene explained: “we run 3 data centers and 4000 virtual servers to ensure our videos and photos are always available.” Why is this important? “If an e-commerce site is making $100,000 per day, a 1 second page delay could potentially cost you $2.5 million in lost sales every year.”
Given these high costs, Gene and his team developed a methodology for choosing where to deploy their next site, based on network intelligence from ThousandEyes. He outlined four simple steps:
- Decide which regions are important to you
- Create some tests
- Let simmer
- Review and analyze
We’ll be delving into these steps in our follow up blog post on Gene’s talk.
DigitalOcean
Next up was Luca Salvatore from DigitalOcean, the cloud infrastructure provider focused on simplifying web infrastructure for software developers. “I've spent the last decade of my life proving a slow app isn't due to the network.” DigitalOcean has been growing fast and today, Luca has 5 network engineers looking after half a million virtual servers in 11 data centers. Luca pointed out that their growth to-date is proof of the “explosive growth” of cloud computing.
He had a lot to share about scaling at speed and automation: “we monitor each of our 11 DCs from ~30 cities around the world with ThousandEyes and every new support team member is trained on ThousandEyes to triage customer issues.” Luca’s talk covered a lot of ground, including responding to mega-DDoS attacks, monitoring for latency and packet loss and how they’re approaching SDN. All of this and more in our follow up blog post on Luca’s talk.
Credit Suisse
Last up was from Darrell Westbury, Director of Operational Analytics at Credit Suisse, the global 200, Swiss-based Investment Bank. Darrell starting off by defining Operational Analytics (OA) as “a method of applying big data principles and data analytics to the operations realm.” OA targets non-traditional types of data: machine data, wire data, agent data, synthetic transactions, and human maintained data.
For Darrell and his team, ThousandEyes provides an important source of information: “ThousandEyes provides simulated views of a customer’s experience while interacting with a service.” “We want to see the world from the service down, not the infrastructure up.”
Darrell then walked through a real-world issue. Two of their ISPs had subcontracted to the same provider. Using ThousandEyes, they were able to catch a load balancing issue before it hit production by having an outside-in perspective: “We had the visual data to share and to get our ISPs to solve the issue.” Darrell shared many more insights and examples, which we’ll include in our follow up blog post on Darrell’s talk.
Network Intelligence You Can See
This is just a glimpse into three interesting and relevant talks. We’ll be following up with detailed blog posts, which will include links to slides and recordings. Thanks to each of the speakers for sharing their valuable experiences and in-depth insights. Stay tuned for our next event (coming soon) and feel free to contact us anytime if you have a particularly interesting network problem or issue you would like help with or to discuss.
You can grow your own insights by getting your hands on network intelligence from ThousandEyes by signing up for a free trial.