{"id":594,"date":"2023-09-27T11:00:00","date_gmt":"2023-09-27T11:00:00","guid":{"rendered":"https:\/\/senser.tech\/?p=594"},"modified":"2023-09-27T13:47:55","modified_gmt":"2023-09-27T13:47:55","slug":"announcing-sensers-emergence-from-stealth","status":"publish","type":"post","link":"https:\/\/senser.tech\/announcing-sensers-emergence-from-stealth\/","title":{"rendered":"Announcing Senser\u2019s emergence from stealth"},"content":{"rendered":"\n

I\u2019m thrilled to announce that today Senser emerged from stealth \u2013 unveiling our AIOps platform and announcing a $9.5 million funding round led by Eclipse, with participation by Amdocs and private investors. <\/p>\n\n\n\n

It\u2019s a major vote of confidence in our team and our technology. And we\u2019re grateful to our investors and customers for joining us on this journey.  <\/p>\n\n\n\n

When Yuval, Or, and I founded Senser in the end of 2021, we had a simple mission: to make it easier and faster to troubleshoot performance issues in distributed cloud environments. <\/p>\n\n\n\n

Traditional observability tech imposes massive hidden costs on enterprises. The cost to configure, customize, extend, and maintain the tech is absolutely staggering (by some benchmarks up to 30% of infra expenses). Worst of all, this doesn\u2019t even include the massive cost of troubleshooting. What most companies get is an alert storm that tells you something<\/em> is wrong \u2013 but don\u2019t tell you why or what to do about it. <\/p>\n\n\n\n

We saw this problem firsthand at our previous company, DriveNets. I recall an incident where a customer reported slow response times from their Kubernetes control plane. Investigation uncovered that the node’s storage was nearly full, but what started this cascade? It took our team nearly 96 sleepless hours to discover that a bug was causing worker nodes to overload the control plane by sending logs continuously, rather than just when something changed. <\/p>\n\n\n\n

If we had been able to correlate the number of writes to the hard drive with how fast logs were being written to the network, we would have arrived at the root cause immediately. But without the appropriate tools, it was like looking for a needle in a haystack. <\/p>\n\n\n\n

Traditional observability approaches \u2013 including both commercial vendors and open source alternatives \u2013 are broken in two basic ways:<\/p>\n\n\n\n