2018 has been a memorable journey at TotalCloud. We released several features supporting our vision to create an intelligent cloud platform. As we enter into 2019, future looks promising.
Talking of future, I have always wondered how will the future of tech look like ten years from now?
Not long ago, technologies like cloud computing, Artificial Intelligence (AI), Machine Learning (ML), the blockchain, etc. swept us off our feet with mind-boggling capabilities.
Use of AI, ML, and cloud have positively exploded. We are not far from the day where Alexa can independently build a mobile app for us with just a command of few words. AI and ML models are changing the paradigm of tech today, including cloud computing and management space. It is, however, not yet able to develop systems in this niche with human-level cognizance.
Today, several cloud management services leverage machine learning or AI. As we know it, ML models need large data sets within a defined environment to make analysis and predict outcomes at human-cognizance level.
So, even in the niche space of cloud computing, AI-powered or ML-based solutions work wonders only when there is the availability of large datasets. Like large enterprises, where they collect several petabytes of metric data in an organized fashion — right from a VM’s memory, CPU, disk utilization to security threats. So, crunching large data sets of metrics and predicting an outcome is easier in such a setup.
Unlike large organizations, availability of metric data is less in SMBs. So, use of AI-powered or ML-based solutions do not provide effective outcomes as expected of them.
In such setups, they need to build hyper-complex data models to get effective outcomes, which is a considerable challenge. And the availability of standardized AI/ML model to address the issues in entirety is still a long way to go. This could be one of the reasons why the use of siloed management tools or consoles is still extensive.
One plausible solution to address AI/ML’s over-reliance on data in the cloud world is Behavioural-AI used in real-time strategy (RTS) games — imagine the likes of WarCraft and StarCraft games. Here, predictions and decisions are based more on behavioral analysis of entities.
Each cloud service exhibits a standard behavior. Take for instance Amazon S3 bucket behaves in a certain way, while Kubernetes clusters have their behavior and Google Preemptible VMs have their characteristics. So, the concepts of decision making in RTS games can be methodically applied to predict and model the dynamicity and traits of cloud computing resources.
Further, with standard cloud APIs already available, these predictions can be automated and together leveraged to build the next level of intelligence by tuning the already made ML models.
Here at TotalCloud, we are slowly moving in this direction of building an intelligent cloud platform featuring innate preference of a human mind.
As I look back at 2018, there were several new feature roll-outs.
Earlier versions of TotalCloud did include live resource correlations mapping. But users had to pan and zoom-in several times to view the resources. The current version automatically pans into the Level of Detailing (LoD) and shows all the correlated resources in a single view.
One of the significant needs of cloud users is to understand the security posture at a resource level without losing the context of other elements in the infra. So we built this Security Group View where AWS users, whether at expert-level or a novice, can easily understand their AWS infra through the lens of security.
This View complements the Security Group View and helps you understand the infrastructure at Network level. Tasks like looking for TCP/UDP ports open to the world, misconfigurations, etc. is a breeze with this View.
A cloud user is empowered when he can get layers of data without losing the context of things. Say, for instance, you have found an orphaned S3 bucket in an AZ open to the public. You want to know how much data it contains and its cost simultaneously. To get all these information on AWS console, you need to juggle between several services’ tabs.
With TotalCloud Perspectives, you can get all these three different layers of data in a single view with a few clicks, in seconds.
The below video shows the cost of each VPC in an AWS infrastructure.
During early 2019, we will be rolling out three key features:
This will be a unified zero-steps view of resource inventory and cost data across all regions of AWS. You get a top-level view of the AWS cloud spend mapped with resource inventory simultaneously.
This view will be able to provide AWS tag-schema in real-time. Using this view, you will be able to find resources based on certain tags you have created for your applications, environment, version, etc. at a glance.
AWS VPC flow logs have been extremely useful in detecting security loopholes and network behavior. This upcoming feature will help you monitor all the data packet flow, in real-time, and help pinpoint issues within the architecture much faster.
Plus, there will be other features, such as:
– Support for multi-cloud, where you can manage and monitor resources from several cloud service providers.
– Custom View, where you can build your View that can render customized visualizations of cloud resources at different levels.
– Cloud Builder, which will allow cloud architects to develop their ideal architecture on the cloud with drag and drop facility and make the entire architecture live in minutes.
– Marketplace, which will enable you to integrate 3rd party services to your cloud infrastructure and further help you provide visibility into the entire ecosystem.
Feature by feature, perspective by perspective we plan to fulfill our vision to build that intelligent cloud platform with human cognizance.
With this, I want to wrap 2018 on a high note and welcome 2019 with more zeal.