Microfluidic Biochips Application Model Microfluidic Biochips Biochemical application is modeled using a sequencing graph G(O, E) that is directed, acyclic and polar. Set O captures the operations that need to be performed and the associated execution time. Edge set E captures the dependency constraints. Biochips are replacing conventional biochemical analyzers and are able to integrate on-chip all the necessary functionalities for biochemical analysis using microfluidics. In Flow-based biochips, liquid samples of discrete volume flow in the on-chip channel circuitry. The biochip has two layers (fabricated using soft lithography): flow layer and control layer. The liquid samples are in the flow layer and are manipulated through microvalves created using the air pressurized control layer. By combining these microvalves, more complex units like mixers, micropumps etc. can be built with hundreds of units being accommodated on one single chip. a Pressure source z1 Control layer Flow layer Valve va Fluidic input Source Sink Mix (4) Mix (5) Heat (3) Mix (3) Heat (2) Filter (5) Mix (2) O1 O2 O3 O5 O4 O6 O7 O8 O9 O10 Microvalve Architecture Model Detector Mixer Filter In3 In4 In2 In1 Out3 Out2 Out1 The biochip architecture is modeled as a topology graph that captures the Biochip components (resource constraints) Biochip inlets/outlets and interconnections between components Permissible fluid flow (routing) paths and their associated latencies Biochip: Functional View Biochip Synthesis z2 z3 z4 In1 In2 In3 In4 z5 z7 z8 z11 Out2 Out3 Out1 v1 v2 v3 v4 v6 v5 Detector Filter z10 z12 Mixer z9 z13 z1 z6 v7 Component Library Currently, researchers manually map the applications onto the valves of the chip which is inefficient, time consuming and would not scale for larger, more complex chips. Problem: Synthesize the application onto the given chip architecture minimizing the application completion time and satisfying the resource and dependency constraints. Component Operational Phases Execution Time Mixer Ip1/ Ip2/ Mix/ Op1/ Op2 0.5s Filter Ip/ Filter/ Op1/ Op2 20s Detector Ip/ Detect/ Op 5s Separator Ip1/ Ip2/ Separate/ Op1/ Op2 140s Heater Ip/ Heat/ Op 20ºC/s Allocation and Placement We consider that the biochip architecture is given. The allocation and placement is captured by the topology graph based architecture model. Mixer The rotary mixer is shown here. It has five operational phases. First two phases (Ip1 and Ip2) are used to move the liquid samples into the mixer. In the next phase (Mix) the samples are mixed, followed by two output phases (Op1 and Op2) removing the mixed sample from the mixer. It has a typical execution time of 0.5s, but this can vary depending on the application. The exact time is specified by the application designer while specifying the application model. Binding, Scheduling and Routing Biochip: Schematic View List Scheduling-based binding and scheduling heuristic utilized Since routing latencies are comparable to operation execution times, thus fluid routing (contention aware edge scheduling) is also taken into account (boxes with labels Fx in figure below) along with the operation scheduling (boxes with labels Ox). As an output, we generate the control sequence for a biochip controller for auto executing the application on the specified biochip. Mixer1 Mixer2 Mixer3 Heater1 Filter1 16 23.5 42.5 56 1 2 3 4 5 6 7 8 9 10 11 F1 F4 O1 F10 F9 F2 F5 O3 F14 S1 S2 S3 Input Waste To other components F25-1 O6 F15 F14 Pump F3 F6 O4 F19 F3 F6 O2 F19 O7 F19 O10 F23 F19 Conceptual View O5 O8 Waste Input O9 The proposed model-based approach is expected to reduce human effort, enabling designers to take early design decisions by being able to evaluate their proposed architecture, minimizing the design cycle time and also facilitating programmability and automation. Schematic View Contact: Professor Jan Madsen Email: Jan@imm.dtu.dk