Boundary Delivers Monitoring in the Cloud
Monitoring network and server assets used to be an activity that required administrators to be inside the boundaries of an enterprise. That's no longer the case.
This week, technology startup Boundary debuted its new real-time monitoring-as-a-service offering. The company also announced that it raised $4 million in Series A funding to help fuel its development for a new paradigm for network and application monitoring.
The Boundary service works by way of agents that are installed on enterprise assets. Those agents then report back to the secure Boundary service where administrators can monitor traffic and applications in real-time for performance or bottleneck issues.
Ben Black, Boundary CEO and co-founder, told InternetNews.com that the goal of the service is to easily enable monitoring, especially for those with little experience.
"We want to change the way that monitoring happens," Black said. "When you can look at everything all the time with really no overhead or expectation of expertise, you will see behavior changes."
Black said that Boundary can be deployed on any type of server, but he noted that initially, the company has been focused on Linux deployments using the Ubuntu and CentOS distributions.
From a data collection perspective, Black said that Boundary is not pulling in log data. Rather the agent looks at every packet. The system then constructs a set of records that summarizes all the flows and is then enhanced with geographic and other information. All of that data is then used for queries and aggregation to stream actionable information. Black stressed that unlike a strictly flow based approach, Boundary is looking at every single packet.
As opposed to just simply using the open source Nagios monitoring system, Black noted that Boundary's approach is different.
"The real-time streaming nature is dramatically different from Nagios or Ganglia or any RRD based systems," Black said. "We really do give the second by second view and we're capable having higher data volumes with extremely low latency."
While the Boundary service itself is not entirely open source, Black noted that his company is open sourcing components.
"The vast majority of components we built internally ourselves, but we have been open sourcing components of the system, including our clustering and load balancing," Black said. "We're looking to open source quite a bit of what we're doing."