下载PDF
实例探究 > Hazelcast IMDG accelerates Financial Market Data access For Investment Banking

Hazelcast IMDG accelerates Financial Market Data access For Investment Banking

技术
  • 分析与建模 - 实时分析
  • 应用基础设施与中间件 - 数据交换与集成
  • 应用基础设施与中间件 - 数据库管理和存储
适用行业
  • 金融与保险
适用功能
  • 商业运营
用例
  • 预测性维护
  • 实时定位系统 (RTLS)
  • 远程资产管理
服务
  • 软件设计与工程服务
  • 系统集成
挑战
Market Data systems in investment banking require the ability to onboard new data feeds quickly and efficiently. Traditional relational databases and NoSQL solutions often fall short due to their reliance on disk-based storage, leading to poor application query performance and unacceptable online customer experiences. Integrating multiple market data feeds without impacting existing throughput is critical, but schema changes in RDBMS can introduce narrow maintenance windows and system impacts. Additionally, trading desk systems demand low latency access to data, which is often measured in microseconds. The complexity and high cost of maintaining these data feeds can stifle innovation and lead to duplication of efforts across multiple development teams.
关于客户
National Australia Bank (NAB) is one of the largest financial institutions in Australia, providing a wide range of banking and financial services. NAB was looking for a replacement for their existing Ehcache-based solution to cache session objects for their API gateway. The existing solution was neither scalable nor capable of providing stable uptime. All digital channels, from mobile to Internet Banking, as well as other groups, access backend systems through this API gateway. The gateway provides security, threat perception, and transformation. NAB evaluated Hazelcast IMDG and Redis, ultimately choosing Hazelcast IMDG for its elasticity, scalability, and highly available distributed solution. NAB has two Hazelcast clusters across two data centers, kept in sync using WAN Replication from Hazelcast IMDG Enterprise.
解决方案
Hazelcast IMDG was chosen as the ideal solution for NAB's needs. Hazelcast uses a single standard Java API and provides native language clients for Java, Scala, Node.js, Python, .NET, and C++, as well as an Open Client network Protocol for building any other native language client needed. Hazelcast is a schema-less in-memory data store, approximately 1,000 times faster than RDBMS, achieving query and update times measured in microseconds for high volumes of data. This allows for the quick and easy integration of multiple market data feeds without impacting existing throughput. Hazelcast's unique in-memory management drastically narrows latency volatility to levels unachievable with NoSQL and RDBMS systems. With Hazelcast IMDG, ROI can be achieved through reduced data integration effort, maximization of economies of scale in terms of memory utilization, and ongoing operational costs, both in terms of hardware and software.
运营影响
  • Hazelcast IMDG provided NAB with a scalable and elastic solution, ensuring stable and always-on access to their API gateway.
  • The solution allowed for more concurrent users in the system with quicker access to the gateway.
  • NAB's two Hazelcast clusters across two data centers are kept in sync using WAN Replication from Hazelcast IMDG Enterprise.
数量效益
  • Hazelcast IMDG is approximately 1,000 times faster than RDBMS, achieving query and update times measured in microseconds.
  • NAB's two Hazelcast clusters across two data centers are kept in sync using WAN Replication from Hazelcast IMDG Enterprise.

相关案例.

联系我们

欢迎与我们交流!

* Required
* Required
* Required
* Invalid email address
提交此表单,即表示您同意 IoT ONE 可以与您联系并分享洞察和营销信息。
不,谢谢,我不想收到来自 IoT ONE 的任何营销电子邮件。
提交

Thank you for your message!
We will contact you soon.