Cisco acquire Splunk, State of Observability Report, SRE CON & more

Welcome!

Welcome to the second edition of the Observability 360 newsletter!

Observability Hits the Headlines!

It has been a momentous couple of weeks - with the obvious headline-grabber being Cisco's acquisition of Splunk. On top of that we have a State of Observability report from New Relic, a raft of new products, a walk through on the TIG stack and even a Reddit pile-on!

Event Overload

The conference season is now in full swing and there are so many conferences and events taking place that there isn't really space to cover them all in the newsletter. If you want a fuller picture just point your browser to the new Events Calendar on the Observability 360 web site. If you are in the loop about an Observability-related event that you would like to be listed, then just get in touch via the email or Twitter accounts below.

Feedback

As practitioners in the field, you will know that every good observability system needs a feedback loop. Let us know how we are doing at:

 [email protected]

https://twitter.com/TheObsGuy

NEWS

Cisco to acquire Splunk for $29bn

The big story of the week/month/year has to be the news that Cisco are set to acquire Splunk for a princely $29bn outlay. As the sector has grown, we have seen a number of acquisitions in recent times. Grafana acquired Pyroscope and have now released their continuous profiling product as Grafana Pyroscope. Similarly, DataDog snapped up Timber Technologies for their Vector data pipeline technology.

Cisco and Splunk executives have heavily played up the strategic importance of Splunk's SIEM business as a driver for the deal. As corporates struggle to deal with data overload and an ever-increasing array of security threats, the bet is that Splunk's advanced analytics capabilities will represent a compelling value proposition.

•  Splunk announcement

•  Cisco blog article

•  SDX Central piece on Cybersecurity/Observability convergence

Bleep! Bleep! Observability on the Tech Radar

Global tech consultancy Thoughtworks have published the latest volume of their influential Tech Radar. In the Tools quadrant, Cilium and Thanos both have Trial status, whilst Pixie is given an Assess rating.

In the Languages and Frameworks sector, OpenTelemetry also has Trial status - although I would guess that within the community a lot of practitioners moved to Adopt some time ago.

Mezmo Launch K8S ‘Welcome’ Pipeline

Observability pipelines are becoming an increasingly important tool for enabling teams to carry out ETL-type operations on their data and route it flexibly between different endpoints.

Mezmo, formerly known as LogDNA, released their first general purpose pipelining tool last year and have now released Welcome as a dedicated solution for taming the vast telemetry output generated by Kubernetes installations.

Dylibso SDK brings Observability to WebAssembly

If your developers are running WebAssembly then you may be missing out on important insights. Traditional agents, tracers etc may not capture key telemetry as WebAssembly code is executed in bounded processes in browsers and VM's.

To overcome this potential blind spot, Dylibso have released their Observe Toolkit, which allows you to "easily extract telemetry in a safe and efficient way through the host/guest boundary". Once extracted the telemetry can be transported through your usual ingestion pipeline.

EVENTS

📣 As mentioned in the intro, there is a glut of events at the moment, so we only have space to highlight a few. See the Observability 360 calendar for a fuller listing!

Grafana ObservabilityCon 2023, 14-15 November, London

Grafana obviously need no introduction and their busy schedule of events, meetings and webinars could fill a calendar by itself. This is a paid-for event at a bricks-and-mortar location so you can meet up with Grafana engineers and fellow Observability specialists IRL. Early Bird tickets are great value, so get in quick!

SRE CON EMEA - 10-12th October, Dublin

SRE Con23 Europe/Middle East/Africa is a gathering of engineers who care deeply about site reliability, systems engineering, and working with complex distributed systems at scale. The speakers list is pretty much a who’s who of Big Tech, with engineers from Meta, Google, Cloudflare, New Relic, Amazon, Reddit and more besides. There will surely be a full house for the “That Time I Accidentally DDoS'd My Company“ session.

From the Blogosphere

Infrastructure Monitoring with the TIG Stack

Observability engineering often involves figuring out the right architectural stack to provide an appropriate solution to a particular technical challenge. In this article, Jay Clifford of InfluxDB sets out a hypothetical use case and then uses the TIG (Telegraf, InfluxDB, Grafana) stack to create a solution for network, server and application monitoring.

Netflix - Observability at Scale

The Netflix engineering team are legendary for their focus on availability and continuity. Here is an article from earlier in the year on how they do alerting - with some useful insights into managing high cardinality data streams.

VIDEOS & TUTORIALS

Relabelling in Prometheus

There is nothing like getting your info straight from the horse’s mouth. Here is none other than Prometheus co-founder Julius Volz with a clear and focused discussion of relabelling:

Event-driven Observability with OpenTelemetry

Event-driven architectures present a challenge for Observability platforms as process flow is interrupted by async messaging systems. Hentik Rexed from Dynatrace looks at using span links in OpenTelemetry to rise to the challenge.

SOCIAL

A Reddit Pile-On

A Reddit user proudly posted that they had replaced Splunk with their own Python script. If you can actually pull that off successfully, then mucho kudos is in order. They probably weren’t expecting the pile-on that ensued here:

Some users were sympathetic to the aim of creating a low-cost, homebrew solution. Others berated him for reinventing the wheel. Some of the criticism seems a bit harsh but if you look beyond the froth there are some really good points being made here about the risks vs benefits of DIY solutions.

Open Source + Observability = OpenObservability

The people behind OpenObservability are committed to “harnessing the power of open source to advance observability initiatives for DevOps practitioners around the world”. You can follow them on Twitter/X to keep up with all of their latest YouTube and podcast output.

Docs

New Relic State of Observability Report

”State of…” reports have been a feature of the DevOps landscape for a few years. New Relic have now delivered their survey of the Observability space. One of the interesting findings is the apparent trend towards teams streamlining their toolsets. Amongst the other findings are:

• 45% of organisations spent $500,000 or more on Observability

• by mid-2026, at least 82% expected to deploy each of the 17 different observability capabilities

• 41% of survey respondents had deployed AI Ops - up from 10% in 2022

Community

📣 If you would like to publicise an Observability-related meetup/standup etc then please let us know and we will list it here.

eBPF User Group MeetUp

eBPF is rapidly gaining traction as a low-level observability tool. You can now collaborate with fellow enthusiasts at the eBPF meetup:

https://www.meetup.com/ebpf-user-group/

OpenTelemetry community meetup

The OpenTelemetry community meetup takes place every month, there are also numerous further meetups based on different specialisms. Even if you are not a Contributor it is a great opportunity to get up to speed with the project that is writing the rules for Observability. There are monthly meetups across three geographical meetings. October’s meetings are:

EMEA (Europe, Middle East and Africa): Oct 17, 11AM GMT (Zoom link)

APAC (Asia Pacific): Oct 18, 11AM India ST register to get the Zoom link

AMER (Americas): October 19, 9AM PST (Zoom link)

That’s a wrap!

That’s all for this fortnight’s edition. As the Splunk acquisition shows, it’s an incredibly exciting time for Observability. See you in two weeks!