SaaS Performance Monitoring

Maintain SaaS application performance through network visibility.

The upside of the SaaS adoption has already been felt by most of us. We can collaborate with colleagues instantly from anywhere while our data are uploaded to the cloud seamlessly. This is, without a doubt way more convenient and effective than the old way of sending documents to each other via email and trying to maintain control over the document version history stored on disks with shared access.

The promise of the SaaS adoption is also to reduce the IT resources required to support application delivery. The reality is that the SaaS adoption causes a shift of priorities in the battle to regain control over the quality of mission-critical systems. Since an employee can connect to an app from a mobile phone connected via crowded airport Wi-Fi, NetOps usually have little chance of ensuring a smooth service. But in many cases, they are being held accountable, regardless.

The lack of visibility and lack of understanding of the communication channels between the applications and end-users may cause communication chaos, paralyse company operations and significantly lower the ROI of adopting SaaS.

Knowing whether the fault is within the network between the affected SaaS user or at the SaaS provider's side a key piece of knowledge that cures a lot of headaches and substantially lowers the impact of SaaS performance issues.
Flowmon Office 365 Server Response Times

List of devices with the worst Office365 Server Response Times

Find the Cause of SaaS Performance Degradation

Since the hot potato of SaaS performance issues usually ends up in the lap of NetOps, having the right data can significantly help them hasten the remediation process. Typically when SaaS performance degrades, there can be 3 different scenarios:

  • Scenario 1: Local network, device misconfiguration, or the path the data travels between
  • Scenario 2: The SaaS provider’s side
  • Scenario 3: Scenario 1 and Scenario 2 are happening simultaneously
SaaS Scheme

With the right data, the responsible staff can choose the correct course of action in each scenario:

  • Scenario 1: Remediate insufficiencies the local network might have. If there are none, contact the ISP and ask for remediation or a different subscription plan
  • Scenario 2: Contact the SaaS provider and ask for remediation or a different subscription plan
  • Scenario 3: Follow the steps described for the 2 aforementioned scenarios
To distinguish between these scenarios, Flowmon uses a non-intrusive method of network performance monitoring by collecting Netflow/IPFIX data. The solution provides network performance metrics from both the local side and the SaaS provider’s side.

In short, Flowmon outright tells you which of these 3 scenarios is your case.
Flowmon Usage of Office365 Components

Usage of Office365 components by volume of traffic
Network Performance Metrics for the Entire Path the SaaS Data Travel

To distinguish between these 3 scenarios, Flowmon measures the round trip time (the time it takes a packet of a TCP handshake to travel from the local network to the SaaS provider and back), server response time (the time it takes for the SaaS provider’s server to respond), jitter (variance of delays between packets) and volumetric statistics (the volume of data that travels at a given moment).

These metrics typically help to distinguish between the 3 scenarios in the following way: 

  • Scenario 1: Round trip time value is above accepted norm. Server response time is within accepted norm
  • Scenario 2: Round trip time value is within norm, server response time value is above accepted norm
  • ​Scenario 3: Both round trip time and server response time values are above accepted norm

These combinations of values help to diagnose where the fault lies. Additionally, volumetric statistics can help to diagnose whether the capacity of the local link can carry the SaaS load. Jitter can help to reveal unstable network connections.