01
Client Profile & Background
A multinational manufacturing company with a production facility in Bekasi, Indonesia, operates a complex supply chain spanning raw material procurement, assembly, and distribution to over 200 retail partners. The company runs an on-premises SAP ECC 6.0 system to manage inventory, production planning, and financials. With 500 concurrent users across departments, the existing server infrastructure—a cluster of Dell PowerEdge R730 servers—was approaching end-of-life and struggling to maintain acceptable performance during peak month-end closing periods. The IT team had already virtualized workloads using VMware vSphere, but the aging hardware limited scalability and introduced reliability risks.
The company's growth strategy included expanding to a second production line, which would increase transaction volumes by 40% and add 150 new users. The existing servers, configured with 8 CPU cores and 64 GB RAM each, were already running at 85% CPU utilization during peak hours. Storage was a mix of SAS HDDs providing only 5,000 IOPS, causing batch jobs to extend beyond the overnight window. The company recognized that a server refresh was essential to support business continuity and future growth.
02
Technical Challenge
The primary challenge was severe performance degradation during month-end closing, where financial consolidation and inventory valuation reports took over 12 hours to complete, missing the business deadline. This was caused by insufficient CPU and RAM for in-memory processing, as well as storage I/O bottlenecks. The existing Dell servers with 8 cores and 64 GB RAM could not handle the parallel processing required for SAP's ABAP work processes. Additionally, the storage subsystem based on 10K RPM SAS drives delivered only 5,000 IOPS, far below the recommended 20,000 IOPS for the database workload.
Another critical issue was backup and recovery. The nightly backup of the SAP database (1.2 TB) took 8 hours, leaving a narrow window before business hours. In the event of a failure, recovery time objective (RTO) was estimated at 24 hours, and recovery point objective (RPO) was 24 hours due to full backups only. This posed a significant business risk. Furthermore, the existing infrastructure lacked redundancy; a single server failure would bring down the entire ERP system, impacting production planning and order fulfillment.
03
Implemented Solution
Intilogy designed a comprehensive server sizing solution based on SAP Quick Sizer results, which recommended a 4-node cluster with 16 cores per node (Intel Xeon Gold 6248) and 256 GB RAM per node, using Lenovo ThinkSystem SR650 servers. The storage was upgraded to all-flash Dell EMC PowerStore 5000, delivering 200,000 IOPS with sub-millisecond latency. The solution included VMware vSphere 7 for virtualization, enabling dynamic resource allocation and high availability via vSphere HA and DRS.
For backup and disaster recovery, we deployed Veeam Backup & Replication with instant recovery capabilities, reducing backup time to 2 hours using changed block tracking and deduplication. The backup target was a Synology RS3617xs NAS with 10GbE connectivity. Additionally, we implemented backup and disaster recovery best practices, including daily incremental backups and weekly full backups, with replication to a secondary site. The network was upgraded to 25GbE using Cisco Nexus 93180YC-FX switches to eliminate bottlenecks.
04
Results & ROI
Post-implementation, the month-end closing process was reduced from 12 hours to just 2.5 hours, a 79% improvement. CPU utilization during peak dropped to 45%, providing headroom for future growth. Database backup time decreased from 8 hours to 1.5 hours, and recovery time objective (RTO) improved from 24 hours to 2 hours, while recovery point objective (RPO) dropped from 24 hours to 1 hour. The all-flash storage delivered consistent 150,000 IOPS, eliminating batch job delays.
The company achieved a 35% reduction in total cost of ownership (TCO) over three years due to lower power consumption, reduced maintenance, and improved staff productivity. The new infrastructure supported the addition of 150 users without any performance degradation. The solution also enabled a 50% faster deployment of new production lines, contributing to a 20% increase in overall equipment effectiveness (OEE). With the robust IT infrastructure in place, the company now has a scalable platform for future digital transformation initiatives, including IoT integration for predictive maintenance.
