Providing support in operations, monitoring, and security reinforcement to ensure stable system operation and drive continuous improvement
NTT WEST Group is a company committed to solving social and regional issues through telecommunications, ICT services, and solutions.
AsiaQuest, together with NTT Business Solutions, Inc. (a group company of NTT WEST; hereinafter referred to as “NTT Business Solutions”), consistently provides operational improvement, optimization, and SRE technical support for the AWS environment that underpins NTT WEST’s business chat system “elgana”. Through this support, infrastructure operation, monitoring, and security enhancements have been strengthened, enabling stable system operation and continuous improvement.
※ SRE (Site Reliability Engineering)
A methodology for ensuring the stable operation of IT systems.
By leveraging monitoring and automation to prevent failures and enable rapid recovery, SRE enhances service reliability while improving operational efficiency.
Background & Purpose
Rapid Response to AWS Operational Challenges
“elgana”, provided by NTT West, is a business-focused chat service.
It is a highly reliable communication platform with numerous implementation achievements, used by customers across a wide range of industries and business types, as well as in the public sector as a mission-critical communication tool in emergencies.
In 2022, the service was rehosted from another cloud provider to AWS. Since then, modernization efforts have continued, further evolving the system as a foundation for business communication.
After the migration, not only was stable operation required, but rapid responses to challenges related to non-functional requirements also became necessary. To achieve this, the support of a partner with advanced AWS expertise was indispensable.
Thus, NTT West engaged AsiaQuest, with a proven track record in AWS co-creation support, to improve and optimize the operation of the “elgana” system.
Role
AsiaQuest supports the customer’s SRE team by working alongside them as a member of the team. Together, we propose and implement improvements to daily operational issues, striving for stable system operation and continuous enhancement.
Key Activities
Investigated the current AWS environment and proposed improvement plans leveraging AWS services.
Conducted regular meetings several times per week during the initial phase, and weekly after migration, to address issues arising from configuration changes.
Improved SIEM on OpenSearch by refining design, selecting relevant data, securing necessary specifications, and configuring access rights for production use.
Investigated causes of cost increases due to changes in usage and implemented several cost optimization measures, achieving significant cost savings.
System Overview
Examples of Operational Support and Improvement Proposals
- 1.Log Analysis
- Adopted SIEM on OpenSearch to visualize information collected from logs and other sources.
- 2.Security Measures
- Detected threat events with GuardDuty, evaluated and validated Amazon Detective, and enabled them across environments.
- 3.Cost Optimization
- Implemented Messenger Application notifications for cost anomaly detection and proposed optimization measures.
- 4.SRE Technical Support
- Supported deployment of AWS resources/applications, incident investigation, various improvement tasks, and updates for middleware and databases.
System Overview
The system configuration of the elgana environment, along with the proposed improvements, is outlined below.
1. Log Analysis
As the log analysis platform, SIEM on OpenSearch was adopted because it could be built quickly and cost-effectively while covering the required features.
In the elgana environment, several hundred gigabytes of logs are generated daily. Importing all logs into OpenSearch would not have been cost-effective or realistic. Therefore, through a proof of concept (PoC), we supported the customer by designing index rotation intervals and shard numbers, reviewing instance specifications, selecting which logs should be ingested for analysis, and creating dashboards.
2. Security Measures
For security alert notifications, based on the customer’s security requirements, it was necessary to detect unauthorized or abnormal activities. To address this, we proposed a mechanism that notifies login events and threat events, enabling the detection of potential security risks.
Furthermore, regarding security incident investigations, when GuardDuty detects a threat event, the customer expresses the need to analyze VPC Flow Logs (particularly for abnormal outbound communications). While the use of OpenSearch was considered, the large log volume and high visualization costs posed challenges. Therefore, we first proposed Amazon Detective, which is easy to configure and provides effective visualization capabilities.
3. Cost Optimization Support
To support cost optimization, we configured notifications in AWS Budgets and Cost Anomaly Detection, enabling early detection of anomalies.
So far, we have identified cost increases in services such as NAT Gateway, S3, CloudTrail, Data Firehose, AWS Config, and CloudFront, investigated their causes, and proposed corresponding countermeasures.
- [Proposal Example 1] Switching from NAT Gateway to VPC Endpoint
- As the frequency of container deployments for batch processing on ECS on Fargate increased, NAT Gateway charges rose significantly. To address this, we proposed switching to VPC Endpoints, which offer lower data transfer costs, thereby achieving a reduction in overall data transfer expenses.
- [Proposal Example 2] Revising AWS Backup Settings for S3
- By changing the AWS Backup settings for S3 buckets from periodic backups to continuous backups, we were able to reduce the number of S3 requests. This, in turn, led to cost savings related to CloudTrail data events and GuardDuty S3 Protection associated with those requests.
4.SRE Operational Support
Since 2025, we have taken over SRE operations previously handled by another vendor, focusing on ensuring stable system operation and driving continuous improvement.
The main responsibilities of the SRE team include:
- Deployment of AWS resources and applications (Terraform, CI/CD, Ansible)
- Incident investigation
- Responding to user inquiries
- Improving operational processes and log management
- Deploying assets and maintaining documentation
- Updating middleware and databases
Outlook for the future
We will continue to enhance the architecture while further accelerating modernization.
In the elgana project, in order to continuously deliver a truly “cutting-edge and innovative elgana” to customers, we plan to accelerate modernization while steadily advancing architectural improvements.
As a co-creation partner supporting the ever-changing and evolving elgana, AsiaQuest will continue engaging in close dialogue with the customer, working together to solve challenges and further enhance service quality.
Testimonial
In the elgana project, our top priority was rapid development and release. As a result, we faced the following challenges in building and operating the service infrastructure:
1.Because we focused heavily on implementation, our monitoring and cost-optimization mechanisms needed to be strengthened.
2.As we accelerated the release cycle, an increasing amount of work became dependent on specific individuals.
Given this situation, AsiaQuest provided practical proposals tailored to our specific needs, and we first began to see tangible results in cost optimization and the enhancement of our monitoring framework.
Furthermore, by consolidating both the build and operational processes under AsiaQuest, we achieved greater standardization through codification and improved visibility. This eliminated dependency on individual expertise and enabled more stable and reliable infrastructure compared to the past.
We strongly feel that AsiaQuest is not merely a service provider, but a true partner that works alongside us to build and evolve the elgana service platform.