The company was founded in August 2015 as the first technology-based logistics company in Southeast Asia with E-commerce as its main business. It handles express delivery, cargo delivery, warehousing, and supply chains. Remarkable success was achieved early on and it became a leading logistics solution provider in the region. With its rapid growth, the customer has not only expanded its business across Indonesia. but also to many other countries, namely Cambodia, Singapore, China, Mexico, Brazil, Saudi Arabia, the UAE, and Egypt, with 23,000 outlets and 350,000 employees worldwide.
To transform the customer's express delivery, cargo delivery, warehousing, and supply chain business into fully automated services using AI, they are investing a substantial amount of resources. They have more than 240 large-scale sorting warehouses, 600 sets of automatic sorters, and 8,000 vehicles. They operate more than 23,000 outlets where instantaneous communication and management is necessary. To tackle this massive undertaking to make the company more agile, the customer is building a comprehensive Hadoop Big Data cluster running the Red Hat Enterprise Linux operating system. To process the enormous amount of data, advanced AI algorithms are required that need massive computing resources. Therefore, the customer will need to purchase a large number of high-performance AI and general-purpose servers to build their infrastructure and meet their rigorous demands.
1. Upgrading conventional system architecture
Technology is advancing at a rapid pace, and all industries are adopting artificial intelligence to enhance their business. Although being a leading logistics support company, the customer is still using a traditional approach for their parcel tracking system and other businesses, which generates an extra degree of latency and deteriorates performance. To meet the tremendous user demands, the customer requires upgrades to the existing system architecture with the utmost advanced AI platform. Thus, to escalate the customer's existing system platform, they are purchasing a large number of highly efficient AI servers and general-purpose servers.
2. Requiring the fastest delivery time and a trustworthy partner
This decacorn is one of the fastest-growing logistics and E-commerce companies in Indonesia. They are dealing with a large number of users at home and abroad. To satisfy the growing business demand, they require an upgrade to their existing platform as quickly as possible, including a large quantity of high-performance AI and general-purpose servers within the minimum time frame. Hence, the delay in the procurement and delivery process may cause interruptions in their business, so fast delivery is crucial. According to the customer, the product lead time must not be more than 6 weeks. The customer is also seeks a trustworthy partner who will ensure smooth operation and after-sales service.
To meet customer demands and effectively solve customer pain points, Inspur proposed a solution with the industry-leading Inspur NF5468M6 AI server equipped with the latest generation Intel Xeon Ice Lake Scalable Processors and NVIDIA A30 24GB GPUs. NF5468M6 was built for extensive AI Inference, video rendering, and multiple AI applications, so the proposal accelerates the performance of the customer's AI-based business solution. Furthermore, Inspur also offered the NF5280M6 flagship servers to build a high-performance and highly efficient platform. As a result, the provided solution greatly maximizes the customer’s demand for computing efficiency and fully satisfies all requirements.
Building high-performance computing clusters to improve the overall efficiency of the AI-based application operation and minimizing the procurement processing time frame.
To meet the high-performance AI computing and reliability requirements, Inspur selected the extremely powerful NF5468M6 AI computing server, which is equipped with 2x Intel Xeon Ice Lake Scalable Processors and 8x NVIDIA A30 GPUs with support for the latest PCIe 4.0. Each A30 GPU provides 5.2 Teraflops of computing capacity based on FP64. A single machine with 8x A30 GPUs has 41.6 TFLOPs of computing power. Inspur also proposed the NF5280M6 flagship server as a good fit for customer application scenarios. With the proposed hardware platform, the customer built the AI-based business solution and established the comprehensive database cluster with BigData architecture. The proposed solution immensely enriched the customer’s system performance and reduced the overall TCO. Inspur's proposal outperforms the other vendors in design, server lead time, and delivery methods by using a global production process, taking only 8 weeks versus the 10-12 weeks needed by other vendors. Based on extensive analysis of the customer’s demands, the product team from Inspur forecasted the 8 week delivery time and successfully fulfilled the order within that time frame, achieving the competitive advantage over other vendors. Inspur also assures the best post-sales service for the customer.
Inspur’s solution enhances the overall computing capacity of the customer's AI architecture which satisfies the rigorous computation demands for the advanced AI algorithm and ensures the smooth operation of the customer's AI-based business platform. The fastest delivery time and prompt after-sales service fulfill the customer's conditions and ensure business continuity. The overall solution improves the development efficiency and significantly reduces the customer's TCO.