Wednesday, October 9, 2019

How to Monitor Hybrid and Multicloud Networks


Organizations today use two or more clouds and distinct tools to monitor environments. Faced with complexity, which approach to take?

 Network monitoring in companies has never been easy. According to Enterprise Management Associates, even before organizations began migrating to the cloud, a typical company used four to ten tools just to monitor and troubleshoot its own networks.
The public cloud adds another complex obstacle to network visibility. Traditional monitoring tools focus on the performance of individual network elements. But today, the age of digital business requires a more holistic view, with the ability to gather and correlate data from diverse cloud environments, using big data analytics and machine learning .



According to a survey by Kentik, currently 40% of organizations consider themselves multicloud users , having two or more cloud service providers. Already a third of companies have a hybrid cloud environment, with at least one cloud service provider and some kind of traditional infrastructure belonging to the third party.

“There are so many different types of data that people collect and analyze on the network - from device metrics to NetFlow to packages and logs to active synthetic monitoring, and no vendor does it all very well. Most don't even try to do all this, ”says Shamus McGillicuddy, EMA Research Director.

As a result, 35% of multi-cloud users have three to five monitoring tools, including log management tools (48%), application performance management tools (40%), open source tools (34%) and performance management tools (25%).

“People tell me they just can't find tools end to end. They have a very good view of the data center, a good view of AWS, a good view of Azure, but they can't put it all together, ”adds McGillicuddy.
For Bob Laliberte, senior analyst at Enterprise Strategy Group, "the environment is getting much more complex." "Therefore, it will be critical to find very sophisticated tools that will allow this complex environment to become simple to manage."

But it's easier said than done. Network professionals often complain that centralized monitoring on existing devices does not increase or provide the visibility needed for cloud and digital age applications. Native cloud monitoring tools, such as Amazon CloudWatch, Azure Monitor, or GCP Stackdriver, are less fragmented and can observe all layers of infrastructure and applications, but some users find that cloud tools often lack sufficient resources.

So far, no vendor has come up with a “comprehensive” monitoring solution, and no solution should be expected anytime soon because of the vast differences between networks. Fortunately, however, there are ways to reduce these differences and get better performance.

Visibility islands


In a hybrid cloud environment, “you will always have islands of visibility. The important thing is to look for opportunities to integrate these islands, ”says McGillicuddy.

One of the most valuable data sources for a network monitoring tool is a management system API used to extract data from other platforms, whether from AWS, or from an IT service management platform such as ServiceNow.

“If you try to put these things together, you'll need a network monitoring vendor that has a very modern API in the tool that allows you to access things like custom data collection, tool customization, and the ability to create new dashboards that let you see the cloud. the way you want it, "explains the expert.

On the plus side, most new vendors have a good API. “Infrastructure teams can have an advantage with some of the legacy tools that are currently expanding into native cloud environments,” says Laliberte. Toolkits such as Riverbed, which integrate SNMP research, streaming, and packet capture to gain an enterprise view of the network in hybrid cloud environments, and SolarWinds advanced network monitoring for local, hybrid, and cloud infrastructure, “offer the opportunity. to link the solutions. "

Many traditional network monitoring tools, however, were slow to take a cloud approach. About 74% of network managers surveyed by EMA say a network management tool has failed to meet public cloud requirements. Among network managers, 28% said this failure was due to vendor inaction or lack of a cloud support roadmap.

“I think we'll get to where all vendors will be 'good' at incorporating some cloud tips with their tools - but I think you'll never see a time when there is true parity,” says McGillicuddy.

Cloud service providers are making progress


For native and multicloud environments, "cloud providers are starting to provide slightly more consistent access to tools to monitor networks that cross their perimeters," says Gregg Siegfried, director of research, cloud operations and IT at Gartner.
According to Kentik's survey, despite progress, many cloud users are still unaware of or taking advantage of some of the existing monitoring capabilities. For example, over half of AWS users surveyed say they are using AWS-provided cloud-specific monitoring tools, such as flow logs.

“Generally, I recommend that customers first try cloud provider tools and native cloud tools before spending time and money with others,” Siegfried adds, “but there is a delta between the visibility you get from a cloud provider. and the visibility you can get with one of these [complementary] products. ”

Multicloud Monitoring

 
New tools have emerged that combine monitoring in multi-cloud environments. The important types of features in these tools are adaptability, supporting collaboration with product development and other infrastructure teams, and integrating data from multiple sources. Some of these platforms include ThousandEyes, Kentik, and APM tools, such as New Relic and Dynatrace, to name a few pointed out by Siegfried.

In April, Kentik announced integrated support for Microsoft Azure. The company began using streaming data from AWS and Google Cloud Platform late last year. The platform also integrates with other cloud infrastructure data sources such as host-level instrumentation, virtual network devices, and container orchestration.

Last year, Internet monitoring provider ThousandEyes extended its Network Intelligence product to multi-cloud environments. The company pre-provisioned IaaS vantage points, including 15 AWS sites, 25 Azure sites, and 15 GCP sites giving them visibility into the performance of specific cloud providers across multiple regions. The solution also gives IT the ability to measure performance across sites, hybrids, inter and intra-cloud.

While Kentik monitors live traffic, ThousandEyes generates synthetic traffic and then reports what might be happening to a hypothetical network transaction. Interest in active synthetic monitoring solutions has increased over the past three years.

AIOps and Advanced Analytics Platforms


As network monitoring focuses on both data acquisition and troubleshooting, analysts see the emergence of artificial intelligence for IT operations (AIOps) and advanced platforms that perform big data analysis and machine learning to correlate. insights between tools.

"You see some vendors like CA doing this with the large data stack they have built, called Jarvis, which connects to different parts of their tool portfolio to correlate insights between them," says McGillicuddy. “They also tried to make it easier for third parties to extract data to correlate insights. Some specialized vendors can also connect to all your tracking items and correlate everything for you in an easy to see the way. We have seen some indications in our research so far that this is really a good approach. "

this article was originally published on ------- Read More


No comments:

Post a Comment