New Relic vs. Splunk Infrastructure Monitoring

New - relic is well suited if you want to analyse the performance of your services and you want to improve it. Integration with multiple services with same account gives a clear picture of flow of your APIs if you have micro-service architecture. New-relic is less appropriate when you want to do logging of your system. As it does not emits every single calls

Incentivized

Splunk Infrastructure Monitoring is well suited for any complicated environment where you have apps and servers across multiple clouds and platforms and products. If you have a data centre where all your apps and servers are in one single network, you could probably get away with older solutions. But for any modern, complex, hybrid-cloud microservices environment, Splunk Infrastructure Monitoring is a must-have.

Incentivized

• gives us an monitoring of all our underlying servers and also we can configure some alerts upon them like CPU and memory alerts.
• Kubernetes cluster monitoring with new relic for EKS gives us and minute details of our cluster utilisation like node usage, pods memory request and limits
• Network traceability for each and every request with response time analysis is great we can trace which component is responsible for generating response delay
• log managements of the logs the infrastructure is generating we can view logs through there only

Incentivized

• SignalFX handles historical metric aggregation exceptionally well, providing a multifaceted approach to event detection based on anomalies.
• SignalFX's cost is incredibly flexible with their pricing model of DPM (data-point per minute) vs the traditional "per host" model that most monitoring SaaS use.
• SignalFX support is responsive and knowledgeable, very eager to help solve your immediate problems.
• SignalFX integrations is vast and constantly growing, making adoption easy even when multiple different open-source technologies are used in your stack.

Incentivized

• I would like to see sort of simulator inside the user interface, that way we can send requests directly from it to test some configuration instead of setting up a test environment in our end.
• It would be nice if the data ingestion can be filtered by APM's. That way we can know which application is ingested most data.
• It would be nice if we could ingest logs (apache, system logs, and other logs) and correlate them with the APM.

Incentivized

• Better integration with our clients (native mobile clients SDKs will be great).
• A way to easily tag filters and move them across metrics/formula.
• Alert system to easy false [alarm] and hard to configure.
• Be able to have group more dimensions (and have more values on each).

Incentivized

The only issue that we have had with New Relic is that the price might be a little expensive for smaller companies. The amount of data you store in New Relic impacts the cost, and can get away from you if you don't work closely with the vendor. Overall though the application is top notch.

Incentivized

Good: Stable system with low error rate Easy to use for simple use cases Bad: UI is not very clear for complex usage Mobile view (when logged in from phone) is bad No library for .net

Incentivized

As an engineer, New Relic has been very quick and easy for me to pick up/install/use. It has been less easy for some of the less technical-minded folks in our organization and their UI still is inconsistent multiple years after refactoring their platform to be New Relic One.

Incentivized

I find that learning the interface can take some time. We need a better show-and-tell on how the Teams pages, Dashboard Groups, Dashboards and charts delay. Advance SignalFlow is sometimes hard to build. Some better samples of advanced SignalFlow would be helpful. For example, Splunk SPL has a vast resource of examples.

Incentivized

Never observed an outage

Incentivized

there are times where browser cache will cause issues that require you to clear your browser before continuing.

Incentivized

The support team has been really helpful and resolved most of the issues on time. However, for a couple of issues, several follow-ups were needed to elicit a reasonable response. The issue was deeply technical and could have been investigated only by their Architects, and bringing them into the ticket took longer than needed

Incentivized

It's better to start by implementing New Relic in one project and test everything. Try to follow best recommended practices and read all the official documentation. Everything seems well tested. Then, start by installing agents to the rest of your projects and keep a close look to all logs and metrics New Relic gives you.

New Relic is the most full-featured offering that we've found, and is incredibly easy to start using with a PHP app. The New Relic agent is installed as a PHP extension so it is able to monitor and track the performance of any PHP app being run by the web server. Other tools required the installation and setup of a PHP dependency at the application level.

Incentivized

They’re not for the same purpose but we’re using NewRelic and Honeycomb for monitoring purposes. NewRelic is used for HTTP client monitoring for system related throughput, error, database and external client monitoring. Honeycomb is used to monitor actual HTTP request/response values. Splunk [Infrastructure Monitoring] is used for real-time application related throughout and error monitoring.

Incentivized

Agent deployment is easily integrated into our workflow. Adding the agent to new servers is quick and painless

Incentivized

• Less time debugging issues or letting issues go unknown
• We know of issues before our customers
• One common tool for logs, apm, infrastructure, and most alerting. Makes for easier developer experience.
• Cost is expensive and is one of highest engineering spends

Incentivized

• Reduced downtime.
• Caused us to get a lot of spam when we redeployed apps and old instances stopped sending metrics. Muting alerts solves this, but people often forget to do it or do it incorrectly.
• Helped us find historical info about instances/apps.

Incentivized