Subscribe
Sign in
Digma Preemptive Observability — Preempt performance and scaling issues in pre-production
Digma is a Preemptive Observability product that identifies performance and scaling issues in pre-production environments and provides a code-level root cause and the severity of the issues, all done automatically and continuously.
91
Replies
Best
WHAT IS DIGMA
Digma's Preemptive Observability 👀 brings an innovative approach for using observability data to preempt issues before they manifest, instead of relying on alerts to fix them after the fact 🤦♂️.
OBSERVABILITY IN THE AGE OF GENAI 🤖
AI code generation is revolutionizing software development, enabling teams to build faster than ever before. But with speed comes risk—without visibility into potential flaws, AI-generated code can introduce performance issues, scalability bottlenecks, and security vulnerabilities that only surface in production. Dealing with a continuous stream of production incidents drags down the team's velocity and creates constant friction for end users.
PREEMPT VS POST MORTEM 🧟
Teams that aim to stop issues early and before they reach production, cannot rely on a postmortem approach using APMs. These tools are built to alert the team to issues and facilitate investigation once the problem already occurred. Digma switches to a preemptive approach. It analyzes the same data as APMs and identifies specific code issues using patterns, nipping them in the bud while still in pre-production. To close the loop, Digma provides AI fix suggestions.
KEY FEATURES
→ Continuously identify code performance, scaling issues, query problems, and other issue types 🐞
→ Cut resolution time by automatically root cause analyzing each issue as well as providing AI-driven fix suggestions ⚡️
→ Prevent breaking changes by highlighting the affected areas and impacted components for each code change and Pull Request
→ Scaling up your application by identifying which areas of your codebase will scale seamlessly and which may create bottlenecks ⚖️
→ Prioritizing technical debt 💰 by assessing existing issues i impact and criticality
→ Using OTEL-based observability with all supported programming languages and platforms. ☕️
→ IDE and code integration: See issues, insights, and analytics within the code itself as well as metrics and traces 🔭
THE PREEMPTIVE OBSERVABILITY ANALYSIS (POA) ENGINE
The Digma Preemptive Observability Analysis (POA) engine introduces an advanced approach to observability by proactively identifying potential issues before they materialize in production. It achieves this by analyzing observability tracing data, even when data volumes are low. Leveraging pattern matching and anomaly detection techniques, Digma’s algorithm extrapolates expected application performance metrics, enabling it to detect deviations or potential problems that have not yet impacted the application. In analyzing the tracing data, Digma pinpoints the issue to the specific responsible code and commits.
We are super excited to bring Digma to more teams and see the kind of impact it makes on your development process. 🙏
Roni & Nir
Digma co-founders
https://digma.ai
Try our live sandbox: https://sandbox.ui.prod.digma.sy...
Looks great! Complementary to our monitoring systems and focus on developers. One questions- Will it work on aws lambda with python?
Hi @roypovar , yes it definitely can. in most performance testing environment engineers are correlating metrics from different version, trying to catch regressions, yet they have no means to identify the root cause of the regressions, which leads to hours or days of troubleshooting, while versions keep updating in production. Also, since it is all about manual defining and recording thresholds, they might not define a metric in a place that experience a regression, an issue that will eventually materialize in production. With Digma performance testing looks completely different: 1. Digma finds issue with no need to pre-define metrics/ thresholds. 2. Per every issue, an RCA is automatically given at the code level. 3. the performance testing environment doesn't have to mirror the production load, as Digma identifies issues by their patterns which doesn't change because of too much or too little load.
Looks great, team. Regarding production, I’m curious about any potential performance impacts and resource consumption. Could you elaborate on that?
Hey @courtney_glymph ! Digma uses your current observability datastream so will not incur any additional performance cost to your application!
This is very exciting to see that you launched here. Super high value product💙
Looks so interesting for code development, will certainly add them to my list to try!
Congratulations! Please can I ask how much it costs and if you have a pay as you go plan? Can you also elaborate on your enterprise plan?
Hi @celine_borsberry1 , please check our pricing page https://digma.ai/pricing, there is a plan for a single team limited by 5 microservices and then for the enterprise plan, it is prices per the amount of microservices, so you can pay as you grow.
very interesting product. Usually performance, scale issues are found in developement cycle and that proves expensive to fix and rerun. How could Digma help reduce cloud cost ?
Can development teams customize the performance baselines and severity classifications for different types of applications, or does Digma use a standardized assessment model?
Congrats! What AI models power Digma’s detection and anomaly analysis? Can we customize the recommendations? also how do you ensure data privacy (GDPR for example)?
Maker
@guy_davidov2 - Hey Guy, we are fully complied with GDPR. Here is a link to our privacy policy page: https://digma.ai/privacy-policy/. Also, in terms of customization: we do with our Digma Enterprise
For multiple teams .
Looks super cool @Digma Preemptive Observability ! What is the implementation process?
@digma @gadi_vardi The implementation process is straightforward, just install our helm chart on your k8s cluster and configure your application to send the traces to Digma. With zero code changes
Does Digma provide benchmarking capabilities to compare performance metrics across different code iterations or branch merges to identify potential regressions early?
I LOVE Sigma. I think for any Java developer this is just a necessary tool in their toolbox! There is no other tool with such a great usefulness/accessibility ratio in the performance monitoring space. With other tools you need to either dig really deep and perform multiple context switches, or there will be just too much noise. I highly recommend Sigma, if you didn't - definitely try it!
Had the pleasure to try and experience Digma first hand; Great product with impressive capabilities and vision. Customer centric & usability like no other!
Hi @valentyn_dudinov , in the pricing page https://digma.ai/pricing/, we mention that we only do annual billing.
Notion