68% of ESG respondents said that their IT environments were more complex in 2018 than the previous two years.
Today’s organizations fundamentally leverage public cloud services to deliver application experiences to end-users across the hybrid cloud. Yet even as cloud has changed the network landscape of IT, the network performance monitoring tools we use have not changed to keep pace. With traditional monitoring tools unable to provide the visibility that is needed by operations, it is time to rethink how we monitor network performance in hybrid clouds. Organizations need new capabilities for network performance monitoring, including:
• Both packet-based and cloud-flow-based network monitoring to get the complete network picture of the hybrid cloud.
• A view integrated across on-premises and multiple cloud providers, or multi-cloud, to enable end-to-end monitoring, from the application to the end-user.
• The ability to validate and measure network bandwidth utilization in the cloud, beyond the provider’s tools, to gain insight into network costs and drive predictable cloud cost models.
Digital transformation has evolved the IT landscape into a hybrid environment, with applications running both on-premises and in the public cloud. With the migration of applications into the public cloud, IT and network operations need to have the same level of visibility across all applications in order to maintain system performance.
Unfortunately, cloud networks and traditional network performance management tools don’t integrate to provide the level of monitoring that organizations need. To eliminate complexity and gain control over new cloud-based environments, network operations needs new tools to see into cloud networks and understand the relationships between traditional and cloud-based applications. The key to fully leveraging cloud and digital transformation is a solution that provides network performance monitoring across all environments, integrating both packet- and flow-based monitoring to provide an end-to-end performance management solution.
With the rise of public cloud usage and new orchestration software, organizations expected that IT operations would become more efficient with reduced complexity. But a research survey conducted by ESG found just the opposite. In our 2018 IT Spending Intentions Survey, 68% of respondents said that their IT environments were more complex in 2018 than they were two years before, and over one in five said that they were significantly more complex than two years ago.
When asked what was behind this increase in IT complexity, the survey respondents noted higher data volumes, increases in the number and type of endpoint devices, increases in the number and type of applications, and the need to integrate on-premises and public cloud.1
This increase in complexity is then driving the improvements that IT and network operations believe are needed to keep pace with these changes. In another ESG survey, respondents were asked what developments were needed for their organizations to improve the management of their hybrid cloud environments. The areas of improvement were led by the needs for better performance monitoring, better integration between on-premises and cloud management, and better integration with existing frameworks (see Figure 1).2
Which of the following developments do you believe need to take place in order to improve the management of your organization’s hybrid cloud environment? (Percent of respondents, N=261, multiple responses accepted)
Source: Enterprise Strategy Group
These results show that the changes brought on by public cloud, including new workloads and infrastructure integration, especially around management and networking, have increased IT operations complexity and have led to the need for a more integrated performance management solution for hybrid cloud.
In order to best monitor the workloads that run on- and off-premises, network operations needs to have the same level of visibility into network performance that they did before. But unfortunately, that level of visibility is not available with most network performance management solutions. In many cases, operations teams are not even able to determine which resources are cloud-based and which are on-premises.
With the different networking layouts and configurations used by cloud-based applications, traditional network monitoring point solutions lack cloud-specific capabilities, and metrics provided by the hyperscale cloud providers are limited in scope, require validation by the enterprise, and stop at the cloud edge. Since they’re unable to access complete views of cloud networks to manage applications—including network bandwidth, traffic flows both within and out of cloud networks, and traffic distribution—network operations teams are blind to key parts of their networks in the cloud at the same time that those networks are increasing in usage. Unable to fully monitor and validate cloud-based network performance, IT has little to no capability to validate the network usage within and to the public cloud and no ability to fully grasp the network traffic costs associated with cloud-based applications.
What network operations needs is a new network performance management technology that integrates both cloud-based network flow and cloud-aware packet monitoring to bring an integrated view of networks and applications and eliminate silos of monitoring.
So why do we need both flow-based cloud monitoring and cloud-aware packet monitoring? Traditional packet-based application performance monitoring provides a wealth of information but it needs to be extended into public cloud services. The on-premises network tools that traditional solutions are based on are often not applicable to cloud-based networks. For a packet-based performance monitoring solution to monitor cloud-based applications, it needs to accept packets not only from traditional network sources but from resources specific to cloud networks, capturing cloud packets across virtual networks. Packet information is critical for determining the application details, network and server response times, as well as the errors that are being generated.
But with the new generation of cloud-based applications, including cloud-native applications, where application monitoring goes beyond just network packet performance, we need flow-based cloud monitoring. When running cloud-based applications, both application performance and costs are spread across the cloud and back on-premises. This drives the need to leverage network flow data to determine:
This level of integrated information allows network operations to analyze the network traffic, from the application all the way down to the end-user, of both on-premises and off-premises networks, creating a complete view of application and network performance across the cloud landscape.
As companies strategically leverage the public cloud to extend their capabilities, they need to build strategic partnerships with hyperscale cloud vendors. But companies that use the public cloud currently lack the full visibility into costs and performance, key drivers to maintaining service levels for users. Early adopters of cloud assumed that the cloud provider was responsible for maintaining performance levels, but as cloud use matures, organizations have learned that providers only supply the infrastructure performance and the end-user service levels are still the responsibility of the enterprise. When problems occur in cloud environments, cloud architects and IT operations need visibility into the end-user experience and network latency, in order to determine responsibility and hold third-party cloud providers accountable.
For organizations to drive value from their strategic cloud partnerships, they must be able to measure, understand, and predict the costs associated with these services. Without transparency into the network traffic flows between cloud instances and across availability zones, as well as the bandwidth consumed by those cloud instances, companies will risk:
Overprovisioning to ensure performance levels.
Unexpected billing charges, due to unpredictable and higher-than-expected metered expenses.
Performance issues that not only affect end-users but drive costs (e.g., apps with chatty network operations, poorly designed architectures that drive metered, external bandwidth usage).
If organizations are given the network and cost transparency required in hybrid cloud environments, they can strategically plan their cloud usage as a component of the application delivery infrastructure, enabling right-sized performance level expectations and reducing unpredictable overall cloud costs.
A unified network performance monitoring solution, with both flow-based and packet-based monitoring, is available with Riverbed SteelCentral. SteelCentral’s network performance monitoring is a key component of its comprehensive digital experience management strategy for the cloud, which spans application, end-user experience, and network monitoring. Specifically, SteelCentral enables network operations to:
• Full visibility into where applications are located and how they communicate on the network, regardless of whether they are on-premises and off-premises, even between cloud networks, eliminating the need for multiple monitoring tools and silos of management.
• The same level of performance monitoring both in the cloud and on-premises, so that existing workflows and process can be extended across the hybrid cloud, reducing complexity.
• A level of insight into network consumption that is not possible with the native tools provided by cloud providers. Multi-cloud visibility also provides the capability to aggregate and optimize traffic across multiple cloud vendors, breaking through vendor management silos.
• The ability to monitor and validate the full network utilization of applications across the hybrid cloud, which is critical for understanding the full costs of the cloud, those which are incurred both inside and outside the cloud providers’ networks.
In the end, to maintain the same level of application and network performance as traditional on-premises environments, IT and network operations need to rethink how they monitor their networks in the cloud age. The increasing complexity of IT is forcing operations teams to introduce new solutions in order to provide the necessary new capabilities to manage cloud-based applications. The combination of packet-based monitoring with flow-based cloud data monitoring delivers the end-to-end network performance management solution needed for the digitally transformed businesses of today.
Riverbed SteelCentral provides unified networking monitoring, empowering organizations with key features to gain better performance management across their hybrid clouds. With expanded visibility across clouds, IT operations can get the most efficient and cost-effective use out of their public cloud and on-premises networks. Companies that are looking for an integrated network monitoring solution, as discussed in this paper, should consider the Riverbed SteelCentral solutions.
1 Source: ESG Master Survey Results, 2018 IT Spending Intentions Survey, Dec 2017.
2 Source: ESG Brief, On-premises Infrastructure Is the Key to Hybrid Cloud, Jun 2017.
This ESG Solution Showcase was commissioned by Riverbed and is distributed under license from ESG.
All trademark names are property of their respective companies. Information contained in this publication has been obtained by sources The Enterprise Strategy Group (ESG) considers to be reliable but is not warranted by ESG. This publication may contain opinions of ESG, which are subject to change. This publication is copyrighted by The Enterprise Strategy Group, Inc. Any reproduction or redistribution of this publication, in whole or in part, whether in hard-copy format, electronically, or otherwise to persons not authorized to receive it, without the express consent of The Enterprise Strategy Group, Inc., is in violation of U.S. copyright law and will be subject to an action for civil damages and, if applicable, criminal prosecution. Should you have any questions, please contact ESG Client Relations at 508.482.0188.