Clutch spoke with David Hickman, the Vice President of Delivery and Engineering at Menlo Technologies, about the comparison between Amazon Web Services and Microsoft Azure – an important discussion for businesses consider leading Cloud platforms.
Introduce your business and what you do there.
Menlo Technologies is headquartered in Santa Clara, California, and has a delivery and engineering office in Hyderabad, India. We have clients around the US. I am the Vice President of Delivery and Engineering for Menlo.
Which cloud platforms are you most familiar with?
We've worked with many platforms over the years. We're currently an Amazon Web Services [AWS] and Microsoft Azure partner. We build applications using these platforms, and also integrate and link into many other cloud sites with various APIs.
What business challenge were you trying to address by using cloud technology?
From a business end-user standpoint, the perception is that things can be done quicker while operating on an Opex [operational expense] cost. Larger companies can achieve the same goals without the normal delays of going through an IT staff. The decision 5 years ago was in terms of Opex versus Capex [capital expense], but currently, even the vendors themselves have stopped investing in on-premises solutions. Oracle has been incentivizing its sales team to recommend nothing but cloud.
Dell Boomi is another strong partner of ours in the integrations space. We've partnered with them because clients like the software and Dell releases features quickly.
Software architecture is currently geared towards microservices and smaller, autonomous consumable units. This allows users to make changes quicker, without the risk of creating major defects in the overall product. Vendors can have bi-monthly or monthly releases, and add cool new features that benefit users right away.
It used to be a matter of saving cost by reducing the hands-on architecture IT team, but we're starting to see better overall benefits for the software because of its ability to be packed in a more agile, nimble fashion.
Menlo Technologies builds cloud-based internet-of-things [IoT] solutions using Azure and AWS, communicating with other services and APIs on the market, for companies that exist 100% in the cloud. Some of our more enterprise clients have on-premises solutions for particular reasons, but the majority is cloud-based and the trend is 'how do I get there'.
What is your recommendation for cloud implementation and migration?
We usually engage in an analysis stage, and the results will depend. For example, when migrating an on-premises SAP, the client won't benefit greatly by moving to the cloud. They might move to a managed service, and not have a local data center, but picking up a large ERP [enterprise resource planning] software and moving it to the cloud is a large effort. The software might be different and could require a rewrite.
In the case of smaller systems, the migration will be easier to manage, and the cloud will offer computing benefits as well. In some situations, the processing power of a data center will be underutilized; when pushing the platform to the cloud, the company will only use what it needs to use through on-demand services like Functions for Azure, or AWS Lambda. Services are spun up as needed and are priced per hour (or less) of computing usage. Ideally, there is a bigger bang for the buck earlier, but there are other risk factors to be weighed in.
A typical large organization will start with development and move to quality assurance. Once there is a comfort level, they will move the whole infrastructure.
Email services like Office 365 are an easier play, relating to the more traditional legacy solutions.
For older reporting architectures, the client may want to move completely to the cloud and adopt Power BI or Tableau through a rip-and-replace strategy.
Do you find cloud as being the more cost-effective option, or does it open the door for more IT spending?
Anyone not finding the cloud to be the more cost-effective option is probably not completely serious about it. Assets on the cloud are more secure than on-premises ones and require a lot less senior talent watching over them. It's clear that moving to the cloud is better, and allows for the adoption of newer technology faster. Companies don't have to worry about a 3-year Capex curve before buying something.
I've run IT for a long time in my career, and have seen a shift in talent, from networking and storage system specialists to technical business analysts or ERP talent. Cloud allows spending on services which are closer to the business.
Is there a specific reason behind utilizing AWS and Azure primarily?
As a company, we go where our clients want to go, and have tried to align with their needs. These are the top cloud systems, so we couldn't go wrong. There are other platforms we work with, but not as much.
Are there any special or unique offerings within any of these platforms?
AWS Glacier is an advanced feature. Their IoT services are there, but a bit discombobulated, compared to those of the Microsoft Azure IoT Hub.
AWS tends to be used more because they've been around longer and have a larger share of the market. AWS offers on-demand pricing called Amazon EC2 Spot Instances, which allows us to build services that can run in a temporary environment, and then receive an email (or electronic trigger) warning us if those services need to be rebuilt. We can tear down and build a system in an AWS environment, and receive a warning if that environment is about to become unavailable. Companies can save a lot of money if they have a big-enough environment, and are mature enough to build in a continuous integration framework.
Microsoft Azure's Power BI is far superior to the new analytics offering from AWS. It's established in the market, has great pricing, and has an embedded streaming analytics for the IoT Hub, allowing for real-time data and integration for statistical modeling.
Azure has Functions that match up with AWS Lambda, so they're comparable in this regard. Both platforms will go back-and-forth and leapfrog each other in terms of pricing and functionalities. They're much more on-par than they were a year ago.
Are there any services which haven't performed up to your expectations, or is there any tool which you'd like to see implemented within each platform?
I'm not familiar with all of their service offerings. A dashboard for all of our services and compute servers would be good. If we resell the platform, from a vendor standpoint, the end-client could then look at and manage their own portal.
Have you had interactions with any of their support team or support resources?
I have only contacted the Microsoft support team so far. Most of the time, like any engineering team, we've looked online trying to figure out answers before bothering a support person. AWS and Azure have support forums, but I don't think that I've ever logged a ticket on either of them.
We have 5 additional questions. For each of these, we ask that you rate AWS and Microsoft Azure on a scale of 1 to 5, with 5 being the best score.
How would you rate them for functionality and available features?
AWS - 4.5
Microsoft Azure - 4.5
How would you rate them for ease of use and ease of implementation?
AWS - 4
Microsoft Azure - 4
How would you rate Microsoft Azure and AWS for support, as in the response of their team, and the helpfulness of available resources online?
I cannot give a rating.
How likely are you to recommend them to a friend or colleague?
AWS - 4.5
Microsoft Azure - 5 - I'd recommend Microsoft Azure a bit more because of the roadmap investment in analytics.
How would you rate Microsoft Azure and AWS for overall satisfaction with the platforms?
AWS - 4.5 - They're a bit more obscure in terms of people that we can talk to, whereas Microsoft Azure has an organization focused more on support. It comes down to the histories of the 2 companies, and how they were started.
Microsoft Azure - 5 - I'm not an end customer, so it is hard to give an overall rating, but both platforms rank high.