Cloud migration is becoming an unavoidable reality for digital enterprise, but legacy monitoring tools weren’t built to fly.
There are two types of operating ratings that define suitable conditions for a pilot to fly an airplane. The basic level is VFR, which in essence limits pilots to fly in cloud-free environments. Then there is IFR, requiring a much greater level of functional knowledge and skills so that a pilot is capable of operating in clouds.
In the case of enterprise IT operational tools for service assurance, a similar system should be put in place to rate operating worthiness with cloud technologies. This is particularly relevant when it comes to the critical tool being used to detect and remediate incidents faster – i.e. the Manager of Managers (MoM) platform.
Legacy event management systems – software like IBM Netcool, CA Spectrum, EMC Smarts, and BMC EM – are effectively “VFR rated,” and lack the abilities required to operate in a cloudy IT environment. As your cloud strategy becomes reality, however, a next-generation MoM that’s “IFR rated” is required. A platform such as Moogsoft’s Incident.MOOG is effectively “IFR rated,” specifically designed to resolve incidents faster in an enterprise IT environment using cloud technologies.
Multiple Routes to Destination: Public, Private or Hybrid?
The element of clarity for the future of enterprise IT is that it’s going to be cloudy. But the cloud still remains as one of the greater mysteries of the IT world. Few IT operational teams agree on what cloud migration is, and even fewer agree on how best to implement it. It’s evident though that everyone wants the benefits that a cloud environment promises.
The many flavors of enterprise cloud adoption are abundant. You’ve probably heard of SaaS, Iaas, and PaaS. Now, overlay that with Public, Private, and Hybrid. Next, throw in the concept of Managed Cloud. The resulting potpourri means that there is a distinct scent of cloud emanating from practically every type of enterprise out there.
Given the diversity, one needs to define a few terms on what enterprise cloud means. In short, there are three types of clouds:
A. Public Cloud – These are multi-tenant solutions generally housed off-premise. Because they are hosted by a third party, and can scale up or down to meet fluctuating demand, the public cloud is great for CRM, e-mail, and web portals.
B. Private Cloud – Unlike public clouds, private clouds do not share disk space with other tenants, making private clouds more suitable for demanding workloads like e-commerce, or an SQL Server database. You’ll pay more for the privacy provided by private clouds, but it’s certainly worth the tradeoff.
C. Hybrid Cloud – This is a combination of public and private clouds, ideally operating together as a cohesive unit. With a hybrid cloud, you can reap the cost savings of the public cloud, while still being able to use the private cloud for resource-hungry or highly secure app apps that can’t live anywhere else. For example, an SQL Server database might require the security of a private cloud for governance purposes, but it will also need to replicate to and from a CRM system in a public cloud – a hybrid cloud fits the bill perfectly.
Flying High with The Hybrid Cloud
Few enterprises will adopt a 100% cloud environment overnight. The cloud may not have won every battle for outsourcing the datacenter—a significant number of enterprises prefer to build their own—but it’s winning the battle for convergence. Through hybrid cloud innovations, the line between the corporate firewall and the massive scalability of tier 1 hosting providers is becoming a blurred one.
Reports have noted that in 2015, 72% of enterprises will pursue a hybrid cloud strategy. Furthermore, the hybrid cloud market as a whole is to reach $80 billion by 2018, according to Gartner research. Looking a couple years out, enterprise cloud strategy will likely include a move to managed private cloud, with off-premise private environments that are actually managed by someone else.
At Moogsoft, we hear a lot from enterprises wanting to migrate to some variation of a hybrid cloud. Some of our customers are web-scale SaaS vendors who have threaded together multiple public cloud solutions, plus some private cloud to optimize performance for the different components of their distribution applications. Others are more traditional enterprises that know they need to move to the cloud for competitive reasons, and are methodically migrating to cloud along with the agile process flows and organizational-cultural aspects that come with this shift.
A Smooth Landing with an IFR-Rated MoM
With clouds on the horizon for almost every type of enterprise, what does this mean for your incident management strategy, and your plans for a MoM to respond to problems faster and more cost effectively?
A MoM sits on top of the entire IT environment and domain-specific monitoring tools, ingesting event and alarm data feeds across all the domains and layers, telling you first where anomalies are unfolding and where operational resources should be focused. Cloud technologies are great when they work, but when things go wrong they obfuscate what’s really happening, delaying detection and response to service-affecting faults.
Here are five capabilities an IFR-rated MoM (cloud rated, i.e. Incident.MOOG) has over legacy MoM / event manager systems that are VFR-rated (not cloud rated):
- Data-Driven Event Correlation. Cloud technologies are inherently dynamic, adapting and scaling infrastructure configuration practically on the fly to adjust for variations in application workload and performance demands. The ability to clean and contextualize data into situations of clustered alerts has to be data-driven, i.e. Incident.MOOG’s algorithmic approach is agile to keep up with changing configuration, unlike the static approaches that legacy event managers use that are based on hard-coded rules and models.
- Public Cloud Monitoring Integration. Public cloud computing services such as Amazon Web Services, Microsoft Azure, and Google Cloud each provide monitoring portals and web APIs for rich event data covering the public cloud domain. Respectively, these monitoring services are called AWS CloudWatch, Azure Management Portal, and Google Cloud Monitoring. These monitoring feeds can be ingested by Incident.MOOG and correlated with the event data from the other domains.
- Intra-Cloud Visibility. An enterprise’s hybrid cloud is going to be designed in layers. This means that if something goes wrong, the visibility to see into all of the layers at once, and contextualize the anomalies across the layers, is paramount to solving problems quickly in a cloud-based environment. This is one of the inherent values of a using a MoM on top of your IT environment and other monitoring tools, provided that a MoM (like Incident.MOOG) supports modern integrations and open interfaces to ingest data from all the layers.
- Inter-Cloud Visibility. Many enterprises will weave together multiple cloud services (public and private) into a hybrid cloud architecture for cost and performance reasons. The ability to ingest data from different cloud domains and correlate between clouds to increase the visibility when something goes wrong will ultimately help overcome much of the potential complexities associated with operationally supporting a hybrid. Incident.MOOG’s ability to work with both public and private clouds together doesn’t add any constraints on how a hybrid cloud should be configured. This helps IT Ops become more responsive and efficient.
- Open Cloud and Containerization Visibility. VMware has a full technology stack for hybrid cloud services, making it a default solution for many enterprises. But new technologies such as OpenStack for building hybrid clouds, and Docker for containerizing application services, are becoming important pieces of many enterprise’s hybrid cloud strategies. Incident.MOOG already supports OpenStack and Docker technologies, ingesting both their event streams and integrating with their APIs. The ability to detect where in the IP stack something went wrong (e.g. what is causing an OpenStack cell or a Docker container problem) is critical to supporting these services in production at scale.
Run Your Safety Check Before Take Off
Unless you’re a cavalier risk taker, I doubt you’d step on a commercial airliner if you knew that the pilots were rated only for VFR, and not IFR. So why are you still using a legacy MoM / event manager that hasn’t been rated to work in the cloud? Now is the time to start preparing for an IT future that only gets “cloudier”.
Over 80% of enterprises currently have a hybrid cloud strategy in place. With this in mind, operational tool architectures and incident management strategies must be ready to turn this strategy into a successful reality.