Diskominfos Helps Bali Pursue Smart Island Digital Transformation on AWS

When the Bali Provincial Government launched its Smart Island initiative to transform the Indonesian island into a digital province, the Communication, Information, and Statistics Agency of Bali (Diskominfos) knew it needed more scalability than the conventional data centers it was using could offer. At the onset of the COVID-19 pandemic, Diskominfos needed to find a solution that would enable public service employees to work from home without sacrificing efficiency.

Bali is an Indonesian island well known for its attractive volcanic scenery and its tourism. In an effort to maintain the harmony of Balinese nature while still advancing technologically, Diskominfos launched the Smart Island initiative with a mission to consolidate public service data resources into a single aggregated management system accessible to both citizens and government workers from one app.

Before its migration to AWS, Diskominfos stored its data in on-premises data centers, but the electricity that these used could be unreliable or unavailable at times. Outages often lasted 20–40 minutes, and some lasted up to 4 hours. These centers were costly and difficult to manage and scale, and they limited innovation because time was diverted to the data centers’ upkeep. “When there was a problem with the electricity supply, the data centers could not run well, and that affected our public services,” says Ngurah Udiyana, head of the informatics application section of Diskominfos.

The COVID-19 pandemic presented the island with an additional challenge—it became unsafe for employees to work from the office. This led Diskominfos to pursue a solution that enabled public service employees to work from home without sacrificing the productivity of having everyone physically present at work. “We needed to make our public services available to our citizens 24/7,” says Udiyana. By using AWS, Diskominfos gained the necessary infrastructure to host a virtual office solution and the scalability to handle concurrent use from multiple employees within the Bali Provincial Government, including teachers from 147 public high schools throughout the province. The level of attendance and the level of performance of each employee determines how much they are paid each month, and Balinese public high school teachers have been able to continue working anytime and from anywhere.

After trying several cloud service providers, Diskominfos chose AWS because of the customer service, scalability, and integrative solutions. “It’s not just about the technology,” says Udiyana. “We get local support and workshops for all of the services, from cloud basics to implementing automatic scaling. AWS gets it done better than anyone else.”

Diskominfos started out by using Amazon Elastic Compute Cloud (Amazon EC2), which provides secure, resizable compute capacity in the cloud. But when the agency’s system was overwhelmed by multiple employees logging in at once, it quickly added Amazon EC2 Auto Scaling, which helps Diskominfos maintain application availability and enables it to automatically add or remove Amazon EC2 instances according to conditions it defines. And by using Amazon DynamoDB, a fast and flexible key-value and document database service, Diskominfos built an attendance system that can scale seamlessly to handle 19,820 public service employees working both remotely and in the office. Employees check in and out of work using a mobile app, so attendance and performance can be monitored without limiting where employees can work.

Diskominfos uses Amazon Rekognition—which makes it simple to add image and video analysis to applications using proven, highly scalable deep learning technology—alongside Amazon DynamoDB to create seamless digital identity verification at the entrance or exit of workplaces. This has been especially useful during the COVID-19 pandemic. With Bali public service employees splitting time between home and work, Diskominfos can use Amazon Rekognition to accurately verify the identity of employees through facial comparison and provide the right resources for each employee after login. This removes the need for employees to spend time logging their locations. All users are verified and receive communication through a notification system built around Amazon Simple Email Service (Amazon SES), a cost-effective, flexible, scalable email service that enables developers to send mail from within any application. Initially, close to 11,000 employees from 247 offices or schools in the area used the new attendance system; that number scaled to over 19,820 full-time and contract employees in August 2021.

Diskominfos employees greatly benefit from the migration to AWS. Udiyana says, “AWS provides the support we need to understand and implement the technology, saving us time and money and making us more productive.” The agency regularly hosts two types of training sessions on using the AWS infrastructure—one for its developers and one for representatives of its users. Since migrating to AWS, Diskominfos has also seen an estimated 69 percent decrease in monthly costs for its attendance system, enabling it to put those savings toward the next phase of the Smart Island initiative.

Many of Bali’s other critical applications are also built using AWS solutions, including a traditional village census system, a health facility oxygen monitoring system, and an asset management system.

Moving forward in its quest to help Bali become a digital province, Diskominfos is working on a queue system for public health services, hospitals, and more. The service will enable people to register before going to a physical location in the interest of efficiency and safety.

Also on the horizon is Love Bali, a web and mobile application that will give tourists access to information about the island, tourist destinations, and ways to contribute to Bali’s sustainability. Love Bali and the public services queue system will both be built on AWS.

“AWS has impressed us with the resources it offers,” Udiyana says. “It has brought us new technology that we couldn’t find before.”

To learn more, visit https://aws.amazon.com/rekognition/.