Question 2 [10 Points] There is a new computing paradigm called Edge computing, which aims to compute at where the data is originated from, e.g., the consumer's mobile devices. Since the mobile devices usually have limited computational power, researchers began to think of smart ways to partition the neural network between the device and the cloud. 1) Generate prediction models CONV FO 1) Extract layer configurations 2) Predict layer performance 3) Evaluate partition points 4) Partitioned Execution Target Application Deployment Phase Runtime Phase Figure 10: Overview of Neurosurgeon. At deployment, Neurosurgeon generates prediction models for each layer type. During runtime, Neuro- surgeon predicts each layer's latency/energy cost based on the layer's type and configuration, and selects the best partition point based on various dynamic factors. The security problem is that when we transmit those intermediate features to the cloud server, we need to consider the man-in-the-middle attack, who can tap into the features and use Generative Adversarial Networks to restore the private training data from the features. The possibility that the attacker can recover the original data declines monotonically regarding the depth of the neural network. For VGG16, the possibility is conv3-64, conv3-64 (90%); conv3-128, conv3-128(70%); conv3-256, conv3-256, conv3-256 (40%); conv3-512, conv3-512, conv3-512 (20%); conv3-512, conv3-512, conv3-512 (5%), for Software-defined anything (SDX) is a collective term that encapsulates the growing market momentum for improved standards for infrastructure programmability and data center interoperability driven by automation inherent to cloud computing, DevOps and fast infrastructure provisioning. As a collective, SDX also incorporates various initiatives like OpenStack, OpenFlow, the Open Compute Project and Open Rack, which share similar visions. As individual SDX technology silos evolve and consortiums arise, look for emerging standards and bridging capabilities to benefit portfolios, but challenge individual technology suppliers to demonstrate their commitment to true interoperability standards within their specific domains. While openness will always be a claimed vendor objective, different interpretations of SDX definitions may be anything but open. Vendors of SDN (network), SDDC (data center), SDS (storage), and SDI (infrastructure) technologies are all trying to maintain leadership in their respective domains, while deploying SDX initiatives to aid market adjacency plays. So vendors who dominate a sector of the infrastructure may only reluctantly want to abide by standards that have the potential to lower margins and open broader competitive opportunities, even when the consumer will benefit by simplicity, cost reduction and consolidation efficiency. 1. Explain the function or purpose of Software Defined Networking Technologies. Answer => Software-defined networking technology is an approach to network management that enables dynamic, programmatically efficient network configuration in order to improve network performance and monitoring making it more like cloud computing than traditional network management 2. List two or examples of computer technologies which the trend of SDN would improve network services. Answer => Edge computing, lot, remote access 3. List two or examples of computer technologies which the trend of SDN would NOT improve network services. Answer =>