Throughput Requirements. The first anticipated customer segment for the Blockchain is California solar installations, but it is anticipated that the Blockchain’s addressable market will expand to encompass all devices that draw a significant amount of power as these devices could all be potentially incorporated into demand/response systems. Such devices include electric vehicles, thermostats, HVAC systems, and large screen electronics. The size of this device category is expected to be in the billions per year. This growth in addressable market over time implies the blockchain need only process less than 1 transaction per second at first to address solar installations but be able to grow to hundreds of transactions per second when the market expands to other types of distributed energy resource devices. Latency measures how quickly a transaction can be finalized after the transaction is submitted to be processed by the system. Some separate the concept of latency into two measurements: latency, the time it takes to put a transaction on a blockchain, and finality, the time the transaction requester must wait to be sure the transaction has been accepted (e.g.- in Bitcoin, the chain that contains the transaction is the one that miners choose to use going forward). They will be treated the same because the present document’s main use case requires finality. The Blockchain’s use case with perhaps the highest sensitivity to latency is key provisioning. Often keys are provisioned on the manufacturing floor which does not lend much time to wait for a transaction to be processed. However, if transactions can be batched together and be decoupled from the manufacturing synchronization this could reduce latency requirements. Therefore, the present document does not mention a latency requirement The following sections will provide details of the different actors involved, as well as further breakdown into individual use cases and activities.
Throughput Requirements. The time to process a job is a function of a number of parameters including whether background of foreground printing is used, complexity of the print job (print quality, number of passes, color settings), load on the network, the CPU (internal HD, amount of RAM, processor), and I/O channel. Three fundamental time metrics are to be used in measuring performance: o Return to application (RTA) o Click to Start printing o Printing o Click to Clunk (just add click to start and printing times). Return to Application (RTA) for Basic and WorkGroup RTA is defined here as the time from starting the printing process (click) to when the host is ready to continue normal working with application (no noticeable performance degradation due to printing process). The generic goal for RTA time equals 10% of the total printing time (click to start plus printing time). This will apply to Eiffel plot suite provided by HP. HP understands that actual RTA time highly depends on specific plot and RIPping computer performance, so a plot suite with specific RTA times defined for each plot will be provided by HP. Reduced RTA is one of the main differentiators of the WG Solution versus the Basic one. This is achieved thanks to the server based architecture where the ripping of the print if performed OEM SOFTWARE DEVELOPMENT AND LICENSE AGREEMENT No 12/00 on the server and so the client workstation is not loaded. This is because in terms of RTA shall be defined two different goals, one for the typical configuration where the client is running on a different workstation than the server, and another where client is running on the same workstation as the server. RTA: client on different workstation than server Under this configuration the Ripping is not done on the client workstation. The WG solution shall provide a RTA for the client which is equivalent to the RTA there would be if instead of a SW RIP, an embedded PS printer was connected to the client. The goal for Eiffel is to obtain same RTA values as DesignJet 1055 when using the PostScript driver (running on same workstation). ICD will provide target numbers for the different plots included in the Eiffel plot suite. RTA: client on same workstation than server
Throughput Requirements. The OC ABI System throughput requirements cover all three classes of workflows. Contractor Criminal TP-TP 482 Other TP-TP 32 ID Slaps-TP 16 TP-LT 482 KP-PLT 385 LT-TP 97 PLT-KP 39 LT-LT 97 PLT-PLT 39 Tactical TP-TP (2+ Fingers) 680 Tactical DNS TP-TP (5+flat fingers) 482 The Throughput requirements are for concurrent ingest and processing of identification, forensic, and tactical Transactions. The average and peak rates are shown below. The Peak rates are expressed as a percentage of the number of transactions above. The Peak Hour will be made up of these peak periods and the rest of that Peak Hour will consist of average loads – for instance the Peak Hour will have a load of a peak 30 minutes worth of Criminal TP-TP and 30 minutes of the average Criminal TP-TP. Criminal TP-TP 30 minutes 10% Other TP-TP 1 hour 20% ID Slaps-TP 30 minutes 10% Tactical TP-TP 5 minutes 2% TP-LT 30 minutes 20% KP-PLT 1 hour 20% LT-TP (100% penetration) 30 minutes 10% LT-LT (100% penetration) 30 minutes 10% PLT-KP (100% penetration) 30 minutes 10% PLT-PLT (100% penetration) 30 minutes 10%
Throughput Requirements. The time to process a job is a function of a number of parameters including whether background of foreground printing is used, complexity of the print job, load on the network, and I/O channel. Under similar conditions the DRIP is expected to process a job according to the following target times. Three fundamental time metrics are to be used in measuring performance: Return to application (RTA) Click to Start printing Printing Click to Clunk Some orientation of the goals for Printing Time for a D-size media is below, but real values are going to be defined according to a throughput test to run at the HP site: Image QA CAD Fast 150x150, coated media ? 1.47 minutes Best 600x600, coated media 21.57 minutes 9.30 minutes Best 600x600, glossy media 21.57 minutes Table 3: Throughput goals
Throughput Requirements. 40 13.3. COLOR PERFORMANCE GOALS .................................................. 41 13.3.1. Color Accuracy vs. Target ........................................ 42 13.3.2 Color Consistency ................................................ 42 14. Licensing restrictions ......................................................... 42 14.1. DEVICE MODEL CHECK LICENSING ............................................. 42 14.2. JAPANESE FONTS LICENSING ................................................. 43
Throughput Requirements. The volume of data traffic consumed by the 5G and beyond 5G (B5G) use-cases, services and applications is expected to significantly grow in comparison to today’s 4G/LTE generation. A factor of approximately 5-10× is foreseen. The experienced 5G user data rate depends on the targeted application/use case, and ranges from few kbps in case of massive Internet of Things to hundreds of Mbps (up to a peak of several Gbps) in case of broadband access in dense urban areas and indoors [3]. These consumers’ performance requirements should be also reflected and supported in access and transport networks. Evolving from 4G/LTE to 5G network architecture, the main change is that the original single-node baseband functions in 4G/LTE are split between Central Unit (CU), Distributed Unit(s) (DU) and Radio Unit(s) (RU) resulting in a so-called centralized network architecture with functional split (Figure 1). This flexible and efficient architecture can deliver the different service requirements of a wide range of expected 5G use cases and applications [4].
Figure 1 Centralized architecture with functional split Table 1 Throughput requirements for the transport network due to a certain functional split option [4] 3GPP split option 2 6 7-3 7-2 8 (CPRI)
Throughput Requirements. A. Image/Demographic Data Capture Time The uninterrupted start to finish time required for a trained live scan device operator to capture the rolled and plain impression fingerprint image data and to record demographic information must be, on average, no more than ten minutes per transaction. This time requirement excludes the printing process.
B. Data Forwarding The process of forwarding data to the central site must not cause the fingerprint scanning function to be disabled, i.e., once initiated, the process must be transparent to the live scan device operator. Each system must provide the operator with the ability to transmit completed transactions to the central site, local printer, or both.
C. Transmission Status Upon completion of the capture of fingerprint and demographic data, the operator must be given the option of sending to the Central Site, printing locally, or both. A default transmission protocol will be established per device at the time of installation. The status of the transmission process must be made visible to the live scan device operator. Therefore, it must be possible for the operator to determine that a given transmission has been successfully completed, is still in progress, and has experienced error in the transmission and/or that the transmission has failed.
D. Transaction queue The live scan queue must maintain a minimum of 200 transactions prior to automatic purging from the queue. Transactions that are not printed or transmitted (depending on the circumstances) must remain in the queue until printed or transmitted to the Central Site. Each live scan device operator must be able to locally print or retransmit any transaction to the central site as long as the transaction remains on the live scan queue.
Throughput Requirements. 28 5.11.1. RETURN TO APPLICATION (RTA) FOR BASIC AND WORKGROUP ........... 28 5.11.2. CHECK TO START PRINT .......................................... 29 5.11.3. PRINTING TIME ................................................. 29
