When I tried infrastructure management services for PCMag, one of the things I noticed was that there was always something missing. This is not a surprise, since the range of things that the organizations enter into the infrastructure can vary a lot. But a common feature of lack of functionality was about the ability to handle both local infrastructure and cloud infrastructure in a single dashboard. For example, if a utility managed infrastructure that runs in Amazon Web Services (AWS) clouds, it will often not cope with the physical switches at home. Amazon has decided to attack this weakness directly with AWS Systems Manager.
It is of course boundaries. For example, like most other infrastructure management tools, AWS defines "infrastructure" primarily as computing instances and software-defined networks. This is not the control tool you should check to see if your printers need toner. But if you need to handle many virtual machine instances (VM), especially across not only multiple accounts, but also across multiple public clouds, then AWS Systems Manager – which will be delivered as part of your AWS subscription top .
Amazon is for some reason downgrading by AWS Systems Manager. While we were able to get on the phone with an Amazon spokesman to discuss it, we were asked not to use that person's name. When we wondered if this person could only summarize the System Manager, the Amazon spokesman explained that customers would operate safely and securely in scale, map their resources to applications and environments, and they need a diverse set of tools to manage their hybrid cloud environments.
"They also have to deal with complex licensing issues," continued the spokesman, "and [as a result] found it difficult to manage the infrastructure [their]. They want the ability to build custom solutions to accommodate their specific business needs." Amazon is dead with that assessment, since it is the promise public cloud has made since its inception, yet it has always managed to wrinkle with unexpected boundaries and difficulties.
Automation at Scale
The spokesman said that there are some important features offered by AWS Systems Manager, which gives users flexibility to get anywhere else, especially when it comes to AWS virtual infrastructure. One such key feature is that you can use AWS Systems Manager to manage resources on a scale. This is difficult to achieve in a large environment that can have tens of thousands of deposits and many different AWS accounts. Other tools often claim this feature, but have performance or data management issues when the scale rises beyond a certain point.
Equally important, AWS Systems Manager gives users a fair amount of real-time control and visibility in their cloud environment. There is a selection of dashboards, including support for creating custom dashboards, as well as a pre-configured and ready-to-use number. This includes dashboards for hybrid systems and Amazon CloudWatch dashboards.
AWS also pointed out some features that are sure to be crucial to many cloud users, including support for compliance through patch manager and a new focus on security. Furthermore, AWS Systems Manager supports AWS automation, hopefully, this option makes it easier to configure. The fact that it can do all this across a range of shuttle services gives IT administrators real freedom to design hybrid shovels that directly address business needs rather than being restricted due to technical technical reasons.
Not Without Competition
All that is said, this is not an AWS advertisement. First, we have not had the opportunity to test System Manager in the same way as we tested our other infrastructure managers. Secondly, AWS is not alone in offering its customers a wide range of integrated management solutions. For example, IBM has its own hybrid cloud management system, as it claims can do many of the same things AWS says is baked in Systems Manager plus it's integrated with IBM Watson so you can use its artificial intelligence (AI) and machine learning (ML) features that help you automate system administration.
If it's really a bad thing, wait until we can test it in PCMag Labs, but on the surface it's possible that Watson can be a great help, especially when the infrastructure scales to a point that makes other systems suffocating. According to IBM, Watson can also help with untraditional automation, such as helping to enforce corporate guidelines and ensure compliance in a large, multi-cloud environment, although the AWS Systems Manager claims that it can also handle this task.
The third major player in this game is Microsoft, which supports hybrid cloud environments with a variety of tools that will manage Azure cloud servers and communities. However, unlike the AWS Systems Manager, Microsoft adds most of these features to its own product range called Microsoft Systems Center – a powerful set of tools, but one that will cost you extra since it's a platform for itself. While Azure has some basic management tools embedded and also includes add-ons, such as Azure Security Center that can protect the entire Azure environment, it can not do much for infrastructure you've hosted in other clouds. On the other hand, if you invest in Microsoft Systems Center, you have no problem managing your community as it is a key focus of the suite.
Finally, while it is a major player in providing cloud infrastructure, Google is not at the same level as AWS, IBM or Microsoft in terms of infrastructure management features, but it works on it. Google Cloud has partnered with Cisco to develop an open management capacity for local and hybrid clouds, although it's not even a rumor when this can be released. Nevertheless, Google focuses on container-based security and will use Cisco HyperFlex for local management.
A Need for an Open Standard
What's happening here is that it's a step to consolidate cloud and local servers as well as cloud-based instances from other clouds in the case of AWS and IBM. This can certainly be a great help for administrators who need to monitor a large virtual environment, as well as a physical environment that is internal.
The fact that AWS and apparently IBM will also let you install their agents in cases that run in competitors' clouds are not something that either advertises, but if you need to manage such a varied environment, this can easily be an opportunity to to make it possible.
It would certainly be more useful if the big players would just create an open standard and then collaborate. That way, you do not have to think about ways to install an agent on a competitor's cloud.