Sustainable data storage solutions: embracing the cloud and big data

数据存储（DS）是数字信息系统蓬勃发展的基本基础之一。随着基于云的系统和高容量移动的设备的日益普及，对可持续存储解决方案的需求变得越来越迫切。当前的DS系统将无法处理数据创建的速率。每天，我们创建2.5 quintillion（1后跟18个零）字节的数据。令人惊讶的是，当今世界上近90%的数据都是在过去两年中创建的。这些数据来自各个方面：社交媒体帖子、图片和视频、电子书、数字音乐、气候数据传感器、金融交易记录、手机GPS和连接信息等等。这就是我们所说的大数据 *。该研究计划面临两个不同的挑战：1）设计高效的DS解决方案，以及2）发明实现容量的代码来写入闪存。这两个领域共同为下一代云存储提供了一个完整的图景，在下一代云存储中，用户/节点将信息本地存储在闪存驱动器上，并具有更高的容量和寿命。在全球范围内，一个高度复杂的网络存储协议提供了一个分布式的信息存储库，预计是可靠的，高效的，可扩展的，低延迟，成本效益，和安全。* 皇后大学信号设计与分析实验室（SDAL）的Yousefi教授团队是图形代码和网络协议设计的世界领导者。分布式网络和无速率编码领域的研究成果将被用来实现我们的目标，在分布式存储研究的建议。闪存的代码设计也将在很大程度上受益于我们以前在代数和随机编码方面的工作和发明。* 这项研究的结果将对加拿大和地球仪的数百万用户产生深远的影响，因为我们改变了当今大数据的存储和共享方式。我们的研究将通过减少所需存储单元的数量和延长许多需要数据存储的电子设备的寿命来节省自然资源和能源。这将通过减少污染对我们的环境产生深远的影响。这些成果为加拿大经济的一个收入为1550亿美元的部门提供了技术优势和高素质人才。该部门是私营部门研发的最大表现者之一，对加拿大国内生产总值的贡献为672亿加元，并雇用了知识密集型劳动力。HQP将在技术创业、研究和开发、演示和技术写作方面接受高度培训。**

Data storage (DS) is one of the basic foundations on which digital information systems thrive. With the ever-increasing popularity of cloud-based systems and high-capacity mobile units, there is critical need for sustainable storage solutions. Current DS systems will not be able to handle the rate at which data is being created. Every day, we create in the order of 2.5 quintillion (1 followed by 18 zeros) bytes of data. Astonishingly, almost 90% of the data in the world today has been created just in the last two years alone. This indicates an exponential growth and will be unsustainable unless disruptive data storage solutions are found. This data comes from everywhere: social media posts, pictures and videos, ebooks, digital music, sensors for climate data, financial transaction records, and cell phone GPS and connection information to name a few. This is what we refer to as big data.***This research program takes on two distinct challenges: 1) designing efficient DS solutions, and, 2) inventing capacity-achieving codes to write on flash memories. Together, these two areas provide a complete picture for the next generation of cloud storage where users/nodes store information locally on flash drives with improved capacity and longevity. Globally, a highly sophisticated network storage protocol provides a distributed information repository which is expected to be reliable, efficient, scalable, low-latency, cost-effective, and secure. ***Prof. Yousefi's group at the Signal Design and Analysis Laboratory (SDAL) at Queen's University is a world leader in the design of graph codes and network protocols. Previous findings in the areas of distributed network and rateless coding will be leveraged to achieve our goals in the distributed storage studies of this proposal. Code design for flash memory will also benefit largely from our previous work and inventions in algebraic and random coding. ***The outcome of this research will have far-reaching implications for millions of users in Canada and around the globe as we transform the way today's big data is stored and shared. Our research will conserve natural resources and energy by reducing the number of storage units needed and prolonging the lifespan of many electronic devices requiring data storage. This will have a profound effect on our environment by reducing pollution. The results provide technological advantages and highly qualified personnel (HQP) for a sector of the Canadian economy with $155 billion in revenues. This sector is one of the largest performers of private sector R&D with a Canadian GDP contribution of $67.2 billion and employs a knowledge-intensive workforce. The HQP will be highly trained in technical entrepreneurship, research and development, presentation and technical writing. **