ABOUT

About AllProbe

Our Mission

Our mission is to build the best distributed IT resources management ecosystem and infrastructures monitoring layer powered by Machine learning AI on top of that. We also strive to synchronize the CMDB model and guidelines with accented IT objects such as hosts & providers titly integrated into IT asset management system(DCIM, and all the IT system parts of the ITIL model) so eventually, support as many use cases as possible and be suited for small to mid businesses as well.

The AllProbe project started as an attempt to create a distributed monitoring system for a comprehensive hosting company in Israel. We embarked on the task since the tools that existed at the time couldn't fulfill our needs of features and expansion. Once completed the monitoring layer we had all the infrastructures elements already in place to be able to extend our service with reasonable effort to managing an entire IT ecosystems, including ITAM (IT Infrastructure Management), DCIM (Data Center Infrastructure Management), Asset Management, Network Devices (aka switches) visualization, IPAM (IP Management) and much more. The monitoring layer on top of the IT ecosystem management allows DevOps teams and others to narrow down a significant portion of the IT management tasks, interfaces, private self-managed systems and service providers to 1 rather easy to use user interface with one viewpoint.

One of the major drawbacks of existing systems was complexity, scalability, the requirement of skilled personnel, the need for many hours of work for basic tasks and many more obstacles. We managed to close the gap of most of those drawbacks.

We are a small team of developers who are committed to this task, and we improve the core components of the systems on a daily basis. We are dedicated to open source foundations, and we believe that ultimately we will publish more codes for the public to enjoy and improve.
We released the first version at the start of 2017 after three years of hard work. This version is built for most of today's needs. We are currently entering a second stage that emphasizes on visualization of data, data analytics and AI (machine learning).

Monitoring architecture explained

Allrpbe is a massively scaled probe-based monitoring system that can scale to infinite probes and Data Centers. One of its main advantages is that it scales by its core components while other products allow scaling with a large amount of effort.

Probe types

The probe server doesn't make a distinction between probes and treats any probe more or less the same. Its divided into probe types in a way that it is relatively easy to plug in any probing needs into the system. The main advantage of doing that can plugin new probe types(e.g., stress tests that are not supported at this point) into the probing grid that handles all the heavy lifting such as history data, rollups based time series for fast charting, events & notifications engines and much more.

Core probe types:
- ICMP
- PORT (UDP,TCP)
- HTTP
- RBL
- SNMP
- JMX

The core probes can be distinguished by 2 main probe types
1 - External check
2 - Internal check

External checks are checks that try to find out a device/host performance by reaching it with a specific protocol and calculate the performance based on the returned value(e.g. fetching a web page and calculate the availability, response code, and response speed)

Goal

We wanted to achieve a system that eventually be able to cover most of the monitoring use cases and that everyone even nonexpert would be able to handle with a minimal amount of learning or reading. We believe that monitoring should be as comfortable as possible.


System explained

The system divided into five core elements:
1 - probers
2 - API
3 - big data & big cache
4 - Client applications(web based UI as a SAAS, smartphone/tablet application)
5 - Background batch processing and analytics


Probers

Probers are the hurt of the system, probe servers come in clusters of at list three servers that compose a quorum in the cluster. A cluster of probes called Data Center, and each Data Center has 1 cluster of probe servers. The probe clusters are completely elastic in a way that growing or shrinking a cluster is natural by adding or removing servers from the cluster, adding probe servers to a Data Center expands the capacity of the cluster to handle a certain amount of probes originating from a Data Center.

Get Started
Sign Up for Free
Signing up is free and takes less than a minute.
Choose the Right Wizard
Easy 2 steps wizards to help you get started with minimum efforts.
Supervise Your Resources
Start master your resources online with reach dashbourd or events based mobile application.