Archive

The Dulin Report

Browsable archive from the WordPress export.

Results (56)

On Amazon Prime Video’s move to a monolith May 14, 2023 One size does not fit all: neither cloud nor on-prem Apr 10, 2023 Comparing AWS SQS, SNS, and Kinesis: A Technical Breakdown for Enterprise Developers Feb 11, 2023 Stop Shakespearizing Sep 16, 2022 Using GNU Make with JavaScript and Node.js to build AWS Lambda functions Sep 4, 2022 Monolithic repository vs a monolith Aug 23, 2022 Keep your caching simple and inexpensive Jun 12, 2022 Java is no longer relevant May 29, 2022 There is no such thing as one grand unified full-stack programming language May 27, 2022 Best practices for building a microservice architecture Apr 25, 2022 TypeScript is a productivity problem in and of itself Apr 20, 2022 In most cases, there is no need for NoSQL Apr 18, 2022 Node.js and Lambda deployment size restrictions Mar 1, 2021 Should we abolish Section 230 ? Feb 1, 2021 TDWI 2019: Architecting Modern Big Data API Ecosystems May 30, 2019 Microsoft acquires Citus Data Jan 26, 2019 Which AWS messaging and queuing service to use? Jan 25, 2019 Using Markov Chain Generator to create Donald Trump's state of union speech Jan 20, 2019 Let’s talk cloud neutrality Sep 17, 2018 A conservative version of Facebook? Aug 30, 2018 TypeScript starts where JavaScript leaves off Aug 2, 2017 Design patterns in TypeScript: Chain of Responsibility Jul 22, 2017 I built an ultimate development environment for iPad Pro. Here is how. Jul 21, 2017 Rather than innovating Walmart bullies their tech vendors to leave AWS Jun 27, 2017 Emails, politics, and common sense Jan 14, 2017 Don't trust your cloud service until you've read the terms Sep 27, 2016 I am addicted to Medium, and I am tempted to move my entire blog to it Sep 9, 2016 What I learned from using Amazon Alexa for a month Sep 7, 2016 Amazon Alexa is eating the retailers alive Jun 22, 2016 In search for the mythical neutrality among top-tier public cloud providers Jun 18, 2016 What can we learn from the last week's salesforce.com outage ? May 15, 2016 Why it makes perfect sense for Dropbox to leave AWS May 7, 2016 Managed IT is not the future of the cloud Apr 9, 2016 JavaScript as the language of the cloud Feb 20, 2016 Our civilization has a single point of failure Dec 16, 2015 Operations costs are the Achille's heel of NoSQL Nov 23, 2015 IT departments must transform in the face of the cloud revolution Nov 9, 2015 Setting Up Cross-Region Replication of AWS RDS for PostgreSQL Sep 12, 2015 Top Ten Differences Between ActiveMQ and Amazon SQS Sep 5, 2015 Ten Questions to Consider Before Choosing Cassandra Aug 8, 2015 The Three Myths About JavaScript Simplicity Jul 10, 2015 Big Data is not all about Hadoop May 30, 2015 Smart IT Departments Own Their Business API and Take Ownership of Data Governance May 13, 2015 Guaranteeing Delivery of Messages with AWS SQS May 9, 2015 We Need a Cloud Version of Cassandra May 7, 2015 Building a Supercomputer in AWS: Is it even worth it ? Apr 13, 2015 Ordered Sets and Logs in Cassandra vs SQL Apr 8, 2015 Exploration of the Software Engineering as a Profession Apr 8, 2015 Finding Unused Elastic Load Balancers Mar 24, 2015 Where AWS Elastic BeanStalk Could be Better Mar 3, 2015 Trying to Replace Cassandra with DynamoDB ? Not so fast Feb 2, 2015 Why I am Tempted to Replace Cassandra With DynamoDB Nov 13, 2014 How We Overcomplicated Web Design Oct 8, 2014 Infrastructure in the cloud vs on-premise Aug 25, 2014 Cassandra: a key puzzle piece in a design for failure Aug 18, 2014 Cassandra: Lessons Learned Jun 6, 2014

IT departments must transform in the face of the cloud revolution

November 9, 2015

This article was originally published on my Cloud Power blog at Computerworld on October 19th, 2015

During the week of October 5th, I had an opportunity to attend the AWS re:Invent 2015 conference in Las Vegas, of all places. As someone who has built an entire career on putting together custom application server software, I should be the last one to get excited about server-less applications. Just like Salesforce’s “no software,” Amazon’s “server-less software” mantra is about a world where application development is not limited by IT, software to install and servers to manage.

Traditionally, in an on-premise environment, putting together an enterprise application involved building out three pieces: the data store, the server and the client. Between the hardware, the software and the infrastructure associated with the aforementioned on-premise application, there were hundreds of thousands of dollars worth of up-front capital expense. The project team was forced to justify their project and their work to the C-suite. CIOs and CFOs were forced to place that initial bet without knowing whether the application will succeed or not.

Of course, computer hardware is just like any other electronic appliance. The washing machine I bought seven years ago continues to wash my clothes at the same rate as it did on the day I bought it. But my family has grown and we now have to do laundry for four people over an entire weekend.

Likewise, as the business grows and business requirements change, the original hardware and software purchased for an on-premise enterprise application will not keep up with the demands. And so project teams are forced to periodically justify increased capital expenditures. The process of explaining one’s existence to the IT department, the CIO and the CFO starts all over again and takes months.

At the Thursday AWS reInvent keynote Werner Vogels maintained a common theme: developers did not get into the software field to keep thinking about infrastructure. Developers have a passion to build applications that are useful and meet user needs. Most CIOs did not get into their fields to keep saying “No” to developers or to create and participate in IT red tape.

To understand the frustration users and developers have with the enterprise IT beadledom, all one needs to do is read commentaries on the rise of Shadow IT, such as this InformationWeek article by Andrew Froehlich:
According to Cisco: “IT departments estimate their companies are using an average of 51 cloud services, when the reality is that 730 cloud services are being used. And this challenge is only going to grow. One year ago, the multiple was seven times, six months ago it was 10 times, today it is 15 times and given the exponential growth of cloud we predict that by the end of this calendar year it will be 20 times or more than 1,000 external cloud services per company.”

[…] Much of the Shadow IT Cisco discovered included Compute services such as Infrastructure-as-a-Service (IaaS) from AWS and Google, as well as multiple storage and backup service providers. On the Software-as-a-Service (SaaS) front, marketing and sales applications such as Salesforce.com dominated.

In a cloud environment like AWS, one or two developers working out of their kitchen or a garage can build and deploy an application that will be useful to millions in a matter of days. Their only upfront cost is their time and effort. Their infrastructure only costs them per unit of utilization. If their idea succeeds and becomes profitable they can scale their application. As new services and technologies become available, the developers can utilize them to cut costs, to offer better performance and more valuable functionality. If their idea fails, they can shut it down and stop paying for the resources.

With great power comes great responsibility. While the need for a dedicated IT department goes away, the need to properly train and monitor developers becomes more important. In an on-premise environment, the IT department acts as the gatekeepers to the use and evolution of the IT infrastructure. The cloud requires a new way of thinking.

Andrew Froehlich writes:
However you ultimately decide to handle the situation, know that the likelihood that Shadow IT can be completely eradicated from enterprise organizations is extremely slim. Rather, the goal for CIOs and IT departments should be to significantly reduce the need for employees to circumvent IT in order to perform their work duties.

Cloud computing democratizes developer and end user productivity at the expense of transparency and IT control. Since developers and users are able to provision and utilize resources as needed, it is easy for costs, overall architecture and security to get out of control. Rather than getting in the way of productivity, however, the IT departments must evolve their role from that of the gatekeepers into that of enablers.