Groundcover is a company that has a goal of reinventing the cloud-native application monitoring domain with eBPF. Pulse 2.0 interviewed groundcover CEO Shahar Azulay to learn more.
Shahar Azulay’s Background
Shahar Azulay founded groundcover in 2021 after working as an ML Manager at Apple. And groundcover co-founder Yechezkel Rabinovich was previously Chief Architect at CyberMDX.
“We both served together in an elite cyber unit in the Israeli Prime Minister’s office where we were exposed to the frustration caused by application performance monitoring solutions,” said Azulay.
Formation Of groundcover
How did the idea for groundcover come together? “Most of my professional background, and Yechezkel’s as well, has been around R&D leadership roles. I have been the end user of application performance monitoring tools for quite some time. Over the years I noticed that APM (Application Performance Monitoring) is a technology that has been around for a decade, yet it has become infeasible for many companies needing application monitoring today: it is hard to integrate, impossible to scale, and offers an expensive full-blown tracing system, or nothing at all,” Azulay explained. “This segment is estimated at over $50 billion, observability is one of the largest and fastest-growing markets in infrastructure software, with teams willing to allocate up to 10% of their IT spend to observability. Giant companies have been leading the APM sector, yet due to growing data volumes and intricate technology stacks, the cost has risen and these solutions have become hard to integrate and demanding to maintain. The result is clear: over 70% of teams do not have an APM tier in place (DevOps Pulse 2022).”
Plus Azulay pointed out that eBPF – which was first introduced in 2014 – enables programs to run directly in an isolated virtual machine inside the Linux kernel and has evolved over the last 36 months to solve new use cases, becoming the next great promise in fields like network infrastructure, security, and observability.
“We decided to build groundcover by leveraging eBPF to provide deep Kubernetes observability, using it to trace any type of event – from network and infrastructure, all the way to services and applications running in the user space,” Azulay added. “We built a system that uses eBPF to collect observability data straight from the Linux kernel while requiring no R&D efforts in the process. Together with a unique edge-compute approach to collect data efficiently, groundcover covers everything yet stores only what matters.”
Favorite Memory
What has been Azulay’s favorite memory working for groundcover? “We recently launched Flora, our new eBPF agent. While building her – we knew we had something golden in our hands. Once completed, we conducted a benchmark that simulated a high volume environment tracking a simple baseline HTTP server application for metrics of CPU and memory consumption before and after the integration of the various observability platforms in question,” Azulay reflected. “Flora demonstrated minimal to zero overhead to the application’s CPU (+9%) and memory (+0%), while Datadog, OpenTelemetry and the Pixie agent inflicted dramatic overhead of 249%, 59% and 32% above the CPU baseline, respectively, and 227%, 27% and 9% above the memory baseline. Additionally, under a limited CPU environment for the monitored application, the overhead added by Datadog, OpenTelemetry and the Pixie agent was also demonstrated to significantly limit the bandwidth for the HTTP server, reducing the volume of handled requests by 71%, 19%, and 12%, respectively, compared to the measured baseline.”
All other solutions except for Flora raised the resource consumption of the application dramatically and in an unexpected manner, potentially causing the application to reach CPU throttling that might degrade its performance or even create an out-of-memory crash (OOM) in a limited environment.
Plus Flora also proved to be highly efficient in the total resources it consumed, making it the most cost-effective solution at a high scale. And when combining the resources consumed by the different agents tested and the overhead measured on the monitored application, Flora consumed a total CPU that was similar to the one used by OpenTelemetry and the Pixie agent. But that was 73% less than the CPU consumed by Datadog. Additionally, Flora consumed 74%, 77%, and 96% less memory than Datadog, OpenTelemetry, and the Pixie agent, respectively.
“I found this to be extremely exciting, as one of the hardest challenges for any modern observability tool is being able to keep up with the ever-growing scale of data – delivering a comprehensive, accurate picture of a system, all while leaving a minimal footprint. With many leading observability solutions dramatically impacting the resource consumption of the applications they are in charge of monitoring – eventually limiting their performance or causing cost surges, engineers are becoming more and more aware of the hidden overhead inflicted by their observability stack,” Azulay emphasized. “Flora is based on two key ideas that allow it to operate at massively high scale. The first, is the use of new, unorthodox eBPF concepts that unlock extremely low overhead data transferring mechanisms. The second is a nearly zero-copy, memory-efficient pipeline which converts observations flowing from the kernel into meaningful outputs. We were able to prove and demonstrate how groundcover works faster and better than our competitors. This is especially mind-blowing to me as we are a team of 20 people creating a technology that surpasses the capability of legendary organizations with thousands of employees.”
Challenges Faced
What are some of the challenges Azulay faced in building the company and has the current macroeconomic climate affected the company?
“In this macroeconomic climate companies are working hard to reduce SaaS costs. What to other vendors could be a difficult headwind is actually an opportunity for groundcover. We set out on this journey with the goal of breaking the volume-based pricing models of other APM players, and are offering teams a way to get more visibility but still manage a responsible budget,” Azulay acknowledged.
Core Products
What are groundcover’s core products and features? “groundcover is a Kubernetes application performance monitoring solution that reinvents the domain with eBPF. Built for modern production environments, it enables teams to instantly monitor everything they build and run in the cloud without compromising on cost, granularity, or scale,” Azulay replied.
Significant Milestones
What have been some of groundcover’s most significant milestones? “At first, the launch of our product is bringing a wind of change into the APM sector which hasn’t had much innovation in the last decade. Secondly, our first round of funding, in the amount of $24.5 million from VCs such as Zeev Ventures, Angular Ventures, Heavybit, and Jibe Ventures showed a vote of confidence in our technology. Since then, the launch of Murre, an open-source tool that helps developers monitor Kubernetes CPU/memory metrics without the need to install anything on the cluster,” Azulay cited. “This tool helps developers get quick access to these metrics without the need to install any 3rd party components on the cluster. And quite close after the launch of Caretta, an open-source tool that helps teams instantly create a visual network map of the services running in their cluster. Caretta leverages eBPF technology to collect data in an efficient way and is equipped with a Grafana Node Graph dashboard to quickly display the dynamic map of the cluster. Caretta maps all service interactions and their traffic rates, leveraging Kubernetes APIs to create a clean and informative map of any K8s cluster which can be used for on-demand granular observability, cost optimization, and security insights, allowing teams to quickly reach impactful insights such as identifying central points of failure or pinpointing security anomalies.”
Also, the most recent milestone was the launch of Flora. All these milestones offer further development of groundcover for the company’s users.
Customer Success Story
Can you share any specific customer success stories? “Sure, our customers are our greatest ambassadors. We work with clientele from a wide spectrum of industries and verticals such as SimilarWeb, Geosite, Dazz, Reco, and more,” answered Azulay. “Lemonade sells renter insurance through a digital platform. They have more than 1 million paying customers. When we met with Lemonade, they told us that they feel that Observability data is key to driving the automation required to maintain resilient systems for about 100 application services. Lemonade grew over the last five years at a compound annual growth rate of 150% and so did the cost of observability in terms of software licensing and developer time for manual instrumentation. They saw Datadog’s billing kept growing and growing as licensing costs were expected to triple this year compared with last year. Lemonade needed a solution that was more cost-effective.”
Lemonade has been using groundcover in production with the company’s DevOps team which supports 200 developers. And they use groundcover’s dashboards displayed on TV screens throughout the DevOps stations to show developers data about their apps in test and development environments.
Not having to instrument applications separately for each programming language has enabled Lemonade’s DevOps team to roll it out, update it and feed data to its packaged dashboards for these environments quickly and easily.
“They see us as a great troubleshooting platform that helps surface problems and show what went wrong before they even know what to look for,” Azulay continued.
Funding
In September 2022, groundcover raised $24.5 million in funding: $4.5 million in seed funding and $20 million in a Series A round. The A series was led by Zeev Ventures and was joined by previous investors Angular Ventures, Heavybit, and Jibe Ventures. The funding is being used for further product development.
Total Addressable Market
What total addressable market (TAM) size is groundcover pursuing? “At over $50 billion, observability is one of the largest and fastest-growing markets in infrastructure software. Customers in this field are willing to allocate anywhere from 1% to 10% of their total infrastructure budget toward observability,” Azulay assessed. “Using conservative estimates for observability as a fraction of IT spending, and assuming IT spending is 5% of revenue (a common industry benchmark), there is a $56 billion TAM for observability vendors today.”
Differentiation From The Competition
What differentiates groundcover from its competition? “Observability is a critical core competency of any team maintaining a complex infrastructure. but why the heck is it SO EXPENSIVE? Almost every engineering team can tell the following tale: There is a bug in production, taking a long time to investigate. The existing logs were insufficient to shed light on the problem and no metric indicated an issue was emerging. After spending hours on troubleshooting this incident, the obvious step of covering this gap from now on is taken, by adding logs and custom metrics that better capture the behavior that could indicate a problem,” Azulay pointed out. “Later on, management notices that the expected monitoring bill for this month is much higher than usual. That logline added created a much higher volume than anticipated. This tale is nothing but reality served daily by legacy APM vendors like Datadog, New Relic, Dynatrace, and others. Those companies make a business out of charging for unpredictable data volumes. These APM vendors are data warehouse companies more than anything else, built to hoard data and make us pay for it, regardless of how valuable it eventually is. They use shady pricing mechanisms to make it harder for users to truly control costs.”
Azulay also mentioned that studies show that less than 1% of this data is useful and ever explored by users. And the other 99% is collected, stored, and processed but never used. Plus paying and storing the 99% you care nothing about so you can reach the 1% you do – is at the heart of the problem that started to form.
Most observability solutions that dominate today’s market were built when engineering teams were mostly working on single-tiered monolith applications. However, the world is moving to cloud-native for speed, scale, and efficiency. And cloud-native environments broadcast massive amounts of data: somewhere between 10 and 100 times more than traditional virtual environments.
“Tracking all that with the same old approach is where data volumes rose fast, and companies were left to pick up the bill. Hence, observability solutions were getting more expensive, but companies were not getting more value,” Azulay suggested. “We built groundcover to be different all the way from how it collects data, to how data is being processed, stored, and priced. We are breaking the infamous cost-to-visibility depth tradeoff. Teams can control their budget and avoid unexpected spikes in cost with a flat, predictable price. We cover everything yet store only what matters. The result is super-granular yet scalable visibility into what’s really happening inside a Kubernetes cluster, at a fraction of the cost that exists today in the market.”
Future Company Goals
What are some of groundcover’s future company goals? “groundcover is on a path to take on the lead of the Kubernetes observability market which is rapidly growing. We want every company to be able to predict their observability bill at the end of the month, have easy access to all their data, and have no protection installation with zero code changes. We aim to make eBPF-based APM accessible and to break the lock-in of legacy APM vendors that exist today,” Azulay concluded.
