TPU Inference Servers for Efficient Edge Data Centers - DOWNLOAD

How Equinix is re-thinking network monitoring for hybrid, multicloud and edge

Categories Edge Computing News  |  Guest Posts
How Equinix is re-thinking network monitoring for hybrid, multicloud and edge

By Zac Smith, Managing Director, Bare Metal at Equinix.

Infrastructure best practices continue to evolve alongside growing investment in the cloud and expansion at the edge, offering enterprises the promise of greater scalability and efficiency, but also new challenges.

One of the critical driving factors for migration out of legacy environments is the value of being proximate to users as well as the myriad applications, ecosystems, and service providers that create outstanding digital experiences.  As a result, digital leaders across all industries are leveraging increasingly complex and distributed infrastructure. This complexity has been enabled by the explosion of “everything” as-a-service: Software-as-a-Service (SaaS), Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), etc. With such diverse hardware and software ecosystems as the building blocks of applications, monitoring becomes vital to maintaining service availability, reliability, performance, and efficiency.

With the maturing of DevOps and orchestration systems like Kubernetes, developers are now able to take advantage of an expanding portfolio of infrastructure options in more places, including specialty appliances in hybrid environments, IoT devices on the factory floor, and bare metal at the metro edge. The result? Applications can be delivered more quickly to users, with a better experience.

That is, if the network doesn’t get in the way!

An effective network monitoring strategy empowers us to head off any issues before they impact customers. That’s why we take digital experience monitoring (DEM) very seriously; it enables us to consistently deliver the high-quality service expected of us. Our DEM strategy is built around specific monitoring solutions that offer our customers the observability and performance capabilities they need.

Network Monitoring in the Cloud

The adoption of cloud, edge and SaaS platforms has allowed enterprises to scale their applications without worrying about infrastructure or application performance and reliability. Edge cloud services, with their strategically distributed data centers, reduce latency to deliver a better end-user experience. But to maintain optimal service delivery while managing demand without straining resources, cloud vendors need to ensure high network performance.

Network monitoring measures network performance and is an important part of observability. Network monitoring solutions rely on device-generated health metrics, event logs, traffic data, and raw packet data to analyze the performance, availability, and reliability of the network and infrastructure. Such tools align the performance objectives of service providers with their business objectives, using business-level analytics.

From multi-cloud, hybrid-cloud to distributed and edge cloud environments, network monitoring tools provide visibility across these environments and measure the impact of network components (routers, switches, etc.) and endpoints (containers, VMs, etc.) on service delivery.

There are three aspects to consider when building an effective network monitoring strategy:

External network monitoring using synthetic monitoring methodology to analyze network peering, BGP, ingress, and egress. This will help:

– Gain visibility into the last mile and internet issues.
– Detect peering issues.
– Adapt to changes in external network conditions.

Internal network monitoring using synthetic monitoring methodology to analyze network bandwidth, throughput, latency, jitter, error rate, and traffic management. This will indicate any issues with:

– Capacity.
– Configuration.
– Third-party services.

DEM that combines synthetic and real user monitoring methodologies to analyze true end-user experience, API performance, and SLA breaches. This helps:

– Detect issues impacting the end-user experience.
– Prioritize incident management based on business outcomes.
– Optimize network and infrastructure for a better end-user experience.

Cloud providers generally offer access to a user portal with basic capabilities such as system health reports and backup/restore options. At the same time, cloud service providers usually lack end-to-end visibility of the network and infrastructure. This is where additional monitoring solutions come into play.

Monitoring tools provide:

– Performance data from different levels of the network and infrastructure, making it easy to evaluate the system’s current state.

– Historical performance data to identify trends and anomalies. This data can be used to improve application delivery.

– Data analytics and AIOps to learn user behavior and correlate performance data trends. This information can be used to create self-healing workflows.

– Alerting when network performance falls below established baselines.

– Dashboards and reports to provide ITOps with a clearer understanding of the system health and performance status.

Optimizing Services With Network Monitoring

At Equinix, we rely on network monitoring to maintain service performance. Our customers demand high performance and operational efficiency so they can deliver consistent service. A lot of our customers are SaaS providers who deliver their applications over the Equinix network and infrastructure. To ensure they provide the best end-user experience, we need to ensure our services are running at peak levels. That’s why we have invested in monitoring solutions such as Catchpoint to stay ahead of any network-impacting incident.

Proactive monitoring allows us to detect and resolve incidents extremely quickly, with minimum service impact. Our network monitoring capabilities enable us to detect performance degradations anywhere – globally – that could have a significant impact on our customers. Consistent and continuous monitoring also makes it easier for us to identify where any network latency might be originating, if a third-party service is responsible for that latency, and if the issue is limited to a specific region or a network type.

What’s more, incident management can become more streamlined and effective with network monitoring. We use our monitoring tool’s alerting capabilities to ensure the right teams are notified should any incident occur. This speeds up troubleshooting and resolution time. Dashboards offer a single-pane visualization of performance across the network and infrastructure that gives the different teams in the organization a quick view of overall performance.

We are also able to provide our customers a snapshot of service health, with advanced data analytics capabilities. This additional visibility into customer experience allows us to optimize our service, based on what works for a specific cloud environment.

A unified and centralized monitoring platform offers a comprehensive view of how the network and the underlying infrastructure resources are performing. Our comprehensive network monitoring strategy is tailored for modern cloud environments and helps Equinix stay ahead of the performance curve. With a well-thought-out monitoring strategy, we can guarantee service uptime for our customers from all over the world.

About the author

Zac Smith is Managing Director, Bare Metal at Equinix.

DISCLAIMER: Guest posts are submitted content. The views expressed in this blog are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Latest News