After considering the queries, clarifications, recommendations and suggestions, the BAC4IGOV hereby decides to include, revise, amend, delete and/or adapt the following provisions:

Item No. Query BAC4IGOV’s Response
1 Part of the technical proposal is to propose equipment that are capable of deploying in DOST. For us to propose the exact equipment, we will need the information for the total rack quantity and server/switch/storage/network assignments per rack. We need this info for us to size up the load density per rack, which then will help us determine the right capacity of the panel boards and branch circuits. Like the format below:

 

RACK 1
Equipment Power Rating (as per name plate loading)
Equipment 1 xx watts
Equipment 2 xx watts
Equipment 3 xx watts
Equipment 4 xx watts
Equipment 5 xx watts
Equipment 6 xx watts
Equipment 7 xx watts
Equipment 8 xx watts
Equipment 9 xx watts
Equipment 10 xx watts
Equipment 11 xx watts
Equipment 12 xx watts
Equipment 13 xx watts

 

 

There are empty racks available for use on both data centers.  The specifications of the racks are as follows:

 

Rack Cabinet:

Dimensions
600mm (W) x 1990mm (H) x 1070mm (D)
42RU
19″ EIA-compliant rack width
With casters and leveling feet

PDUs:
1 PDU per rack cabinet

Dimensions
55mm (W) x 1790mm (H) x 45mm (D)

Power rating
30A maximum input current
24A maximum continuous current
200-240Vac, 60H, 1P

 

Output Receptacles per PDU
36 outlets x C13, 12A max
6 outlets x C19, 16A max

Power Cord
30A max, 3m, NEMA L6-30P input power cord
Each bidder will be given 4 rack cabinets per data center.  If the rack cabinets being offered by DOST-ICTO do not meet the bidder’s requirements, the bidder may opt to provide its own rack cabinets with the condition that it will deliver only 4 rack cabinets per data center. The contractor has free rein on how the equipment will be arranged in each rack cabinet.

2 Commercial Queries:

1.    Is this a pure subscription based? Who will own all of the Infrastructure after Year 1?

2.    Will the Upgrade after Year 1 be Bidded out or just a Change request?

3.    Are the licenses of the Software mentioned in Ref 3.3.1.5 like MS, Redhat, Oracle are not part of the ABC?

4.    On FPF 2, is the Subtotal (as indicated on the table) equivalent to the Total VATex of our cost proposal?

5.    On FPF3, is this a summary of FPF 4 to 6? Are we to design the Activity No. for the particular activity as identified by the Description?

6.    On FPF 4-5, what does the Name means? Is it the same as the description of the activity?

 

1.    No. DOST-ICTO will own all the infrastructure offered in the solution after Year 1.

2.    Whether to tender a new bid or undergo a change request for the Infrastructure upgrade has not been decided yet.

3.    Yes.

4.    All price proposals should be INCLUSIVE of VAT.

5.    Yes, it is the summary of FPF 4-6. Yes, you are to design the Activity No. for the particular activity as identified by the Description.

6.    No. Name refers to the personnel in charge of the activity.

3 After Sales Support Queries

1.    Who will manage and operate the Cloud Infrastructure after Year 1?

2.    For the Marketing and Training, will this be included only in Year 1? What will happen in Year 2 and onwards?

3.    Service Desk (Level 1) Role

a.      Is this a Managed Service or a Staff Augmentation engagement? Who will supervise the L1 resources?

b.      What is the expected extent of role? Are we to solve L1 incidents only based on known resolutions from knowledge base or are we expected to won the incident until it is totally resolved? (L1 role or full-ownership with L1 as the SPOC incidents)

c.      Will incident reporting be included in L1 support? Will this cover L1 issues only or even those escalated to L2/L3?

d.      Will L1 be responsible for collating all Managed Services performance reports and consolidate all of them for reporting to ICTO management?

e.      For service interruption, who will be the primary POC for all clients? Will L1 be the one to make announcement over self-service portal?

f.       What is the expected escalation process from L1 to L2/L3/L4? If issue is unknown do we close the incident ticket and log a problem ticket in the next level’s own problem ticketing system?  (we need to understand fully the expected escalation process)

g.      If we are to support hardware and software alike riding on the cloud, what are the expected technical skills of L1 support?

h.      Will there be a need to station on-site engineers at various client sites? If yes, what will be their scope of work?

i.       Will there be a need to station on-site engineers and/or L1 support inside iGov data center? If yes, what is the scope?

j.       Assumptions to validate:

k.      Post implementation technical briefing complete with support manuals on various services over iGov Cloud will be provided by L2/L3/L4. The objective is to equip L1 with sufficient technical knowledge to perform basic troubleshooting

l.       Every time there is a new service on boarded iGov cloud by L2/L3/L4 they will do technical briefing on the new service with support manual

m.    Any changes on the service catalogue of L2/L3/L4 will be oriented.

n.      L2/L3/L4 will send document to L1 containing the resolution of a problem, and specifics on how it can be resolved next time at L1. This will be recorded in L1’s knowledge base.

o.      What are the security requirements, if any, as to the off-site location of L1 support team (security access badge, secured phone line and VPN access to iGov, secured vpn, etc.)?

p.      Will there be instances that L1 will have to go on-site to quickly address client issues? If yes, any engagement related expenses/ incidentals incurred such as travel (i.e. such as airfare, transportation, and accommodations), representations and marketing spend shall be billed to the winning bidder.

 

1.   The answer to this question depends on a lot of factors. There are no clear plans on this issue until we have a production cloud already working.

2.   Yes. With regard to training and marketing after year 1, there will still be a marketing and training component.  But as to whom will conduct it will really depend on whether we tender for a new bid so that a new contractor can come in, or DOST-ICTO will handle the operations (including marketing and training) by themselves.

3.   Service Desk (Level 1) Role

a.   It’s a managed service.  The contractor will manage the L1 resources.

b.   The contractor is expected to own the incident until it is totally resolved.

c.   Yes. Incident reporting should also cover those escalated to L2/L3.

d.   It depends on your setup. The important thing is we get the consolidated reports regularly.

e.   The primary point of contact for all clients will be the L1 support. We leave it up to the contractor to designate someone who will make announcements in the self-service portal. Ideally, this could be the Operations Director.

f.    In section 3.3.3.4, it states:
The Contractor shall propose both incident and problem resolution support team structure and the escalation procedures for incident and problem resolution. The procedure shall include addressing Business Severity 1 and 2 that are not resolved within the stipulated resolution time and the follow-up action until resolution of incidents and problems.

It is up to the contractor what the process will be as long as it is consulted with the DOST-ICTO.

g.   We refer you again to section 3.3.3.4 and also item 1.11.  Since the contractor will be defining the procedures, and the Technical Support Team Structure, this implies that it is also responsible for defining the expected technical skills of the L1 support.

h.   No, there is no need to station onsite engineers for various client sites.

i.    This can be decided during the negotiation phase of this procurement.

j.    Assumptions to validate:

k.   Yes, this assumption is correct.

l.    Yes, this assumption is correct.

m. Yes, this assumption is correct.

n.   Yes, this assumption is correct.

o.   This can be finalized during the negotiation phase of this procurement.

p.   If L1 support is allowed to go on-site, then costs incurred by the L1 support shall be borne by the contractor.

4 Training Role:

a.    Is the role will include creation of courseware and training materials, and actual training of key users of clients?

b.    Will the Training materials can be in PDF format and can be printed or accessed from a shared drive specific for training to be provided by ICTO?

c.    Training will focus only on how to use self-service portal. Any form of trainings outside of self-service portal will have to undergo change management and may mean another service agreement with ICTO?

d.    Any engagement related expenses/incidentals incurred such as travel (i.e. such as airfare, transportation, and accommodations), representations and marketing spend shall be billed to ICTO?

 

a.    Yes. Section 3.3.4.10 describes the scope of work for training.

b.   The training materials can be in PDF.  Other types of learning media such as training videos and screencasts may be required also.  All these materials should be published online, preferably in a designated section of the self-service portal.

c.    The answer to this question depends on who the contractor is training. If clients of the cloud are being trained, then the self-service portal will be the scope of the training. If DOST-ICTO technical people are being trained, then the entire technical documentation of the GovCloud will be taught to the staff. Please see 3.3.4.12.

d.   No. All expenses related to the conduct of training shall be borne by the contractor.

5 Marketing Role:

a.    Will the Role include design, build and implementation of marketing plans, activities and collaterals related to services hosted in iGov?

b.    Is this Project include Marketing and any form of POC or pre-sales activities will mean Sales engagement rather than marketing and may mean another service agreement ICTO?

c.    Any engagement related expenses/incidentals incurred such as travel (i.e. such as airfare, transportation and accommodations), representations and marketing spend shall be billed to ICTO?

 

a.    The marketing role will include design, build, and implementation of marketing plans, activities and collateral for the GovCloud services only.  This will be carried out by the contractor.

b.    Marketing can take any form e.g. POC, and pre-sales.  Please note that we consider POCs and pre-sales activities as part of marketing.  Furthermore, this should not be considered as another service agreement but as part of the Marketing component of the GovCloud project. Hence, the cost for these expenses should be estimated and included in the bid proposal for the GovCloud project.

c.    No, incidental expenses such as airfare, transportation, accommodations, etc can not be billed to DOST-ICTO.  These expenses should be estimated and included as part the bid proposal for the GovCloud project.

6 Services: On Progress Reporting

1.6.7 The Contractor shall ensure that a customer satisfaction is carried out at least once a calendar year. The format of the survey shall be proposed by the Contractor and agreed to by the DOST-ICTO before the survey is carried out. The survey shall include 60% of the Customer’s user base.

The Contractor shall also carry out the collation of the survey and produce the survey report within four weeks after the completion of the survey.

 

1.6.8 If the customer satisfaction survey indicates that the overall satisfaction for the Customer’s user base is less than 80%, the Contractor must propose a plan to improve the quality of the service within one (1) month of the completion of the survey report. Another satisfaction survey shall be carried out with the Customer’s user base within six (6) months of the first survey.

 

i. is there a required or limitation for communication channels <LANDLINE, MOBILE PHONE, EMAIL, ONLINE MESSANGER, WEBSITE, SERVICE PORTAL, SOCIAL MEDIA , ETC.> to conduct customer satisfaction survey for the 60% of the customer user base?

 

Services: On Service Level Agreements

Service Desk Response Time (Phone, Email)

Description: The SLA measures the response time of the Service Desk in answering calls and emails. Support Calls are inclusive of queries, feedbacks and complaints from customers (i.e. all users of the GovCloud platform)

 

Measurement Methodology:

The clocking of service starts when the call is connected through to the time support call is answered by Service Desk.

 

 

 

 

Call answering rate is calculated as:

(Total Support Calls answered by Service Desk per month) / (Total support calls received per month) x 100%

 

Success Criteria:

Answer 80% of support calls within sic (6) rings

All unanswered calls will be routed to voicemail system

 

ii. What is the projected volume of calls per month?

iii. Is there a specific system that the customer will use to measure the success criteria or this is expected to be available from the contractor?

iv. Is there a specific voicemail system required?

 

Portal Service Provisioning Turn-Around Time (Self-Service)

The SLA applies to the provisioning of virtualized resources for the cloud services ordered by tenant agencies (self-service) through the portal.

 

v. What are the expected features and functionality that needs to be available on the self-service portal requirement?

 

i.       None.

ii.     Estimated emails and calls monthly average service requests: 275 emails average monthly, 35 calls average monthly, 100 requests through chat average monthly. The system to measure success criteria for should be provided by the contractor.

iii.    None.

iv.    See Section 1.4.1.1 for the features and functionality of the self-service portal.

 

7 Technical Queries

1.    Is the 500VM’s capacity for both Sites are active? Or to support Disaster Recovery, do we need to accommodate 1000VM’s per site? Otherwise, provide recommended capacity that will failover to Secondary Site?

2.    As per TOR, ICTO will be providing the Network Connectivity. What will be the available connectivity between two Sites? What is the estimate number of Agencies?

3.    Provider will be the SPOC for every agencies? What is the estimate number of Agencies?

4.    Regarding, GovCloud DNS, may we know how do you want to use the said DNS? Do you need this for Internet browsing or you want us to host records? Please expound further.

 

1.    Yes the 500VMs per site are active.

2.    Network Connectivity between the two data centers is 10 Gbps.

3.    Yes. Estimated number of agencies, and pending vm requests: 33 agencies, 44 vm requests on queue

4.    The DNS is going to be used for both – Internet browsing and hosting records.

 

8 Regarding CV’s, stated in the Instructions to Bidders that the expert should submit a signed written commitment stating that the expert shall work for the project once awarded the contract. We requested for a template, or a guidance as to the contents of the written commitment. For the template of CV’s, please refer to TPF 6. Format of Curriculum Vitae (CV) for Proposed Professional Staff.

 

9 Also stated in the instructions to bidders that a zero rating shall be given to a nominated expert if the expert is proposed for a domestic position but is not a Filipino citizen. We request for clarification as to the meaning of Domestic Position. Domestic Position is when an individual is employed in the Philippines.

 

10 With regards to the Experience of the Firm, maximum of twenty (20) pages introducing the background and general experience of the consultant, including its partner and subcontractors, is the 20 page limit applicable per consultant or will it be 20 pages for the entire consortium? If the winning bidder consists of a consortium, the entire consortium will be only be allowed twenty (20) pages to describe the background information and experience of its consultants.

 

11 1.3.1.42 Back-up to tape; mentioned on service catalogue but no supporting technical details on the back-end. We request for the supporting technical details. The technical details of the Backup to Tape service are actually described in 1.4.1.43.  For your reference, here they are:

Backup to Tape Library shall have the following capabilities:

·         Should support up to 6 x LTO6 drives and 96 drive slots.

·         Provide 8 Gbps native FC connectivity to SAN switches.

·         Partitioning support so that each drive can be configured in a separate partition.

·         Support Barcode reader and mail slot.

·         Shall be rack mountable with redundant power supply.

·         Supplied with software which can predict and prevent failures through early warning and shall also suggest the required service action.

 

12 We request for a list of the pilot agencies. We can provide the list of pilot agencies to the winning bidder during negotiation.
13 We request for a copy of the developed service catalogues. If the intent of this question is to determine how the 500 VMs in the Makati NGDC and the 500 VMs in the Quezon City NGDC are going to be sized, please refer to the table below:

Size vCPU and vRAM Quantity (VMs, Makati) Quantity (VMs, Quezon City)
S 1vCPU, 2GB vRAM 200 200
M 2vCPU, 4GB vRAM 100 100
L 4vCPU, 8GB vRAM 100 100
XL 8vCPU, 16GB vRAM 100 100
14 We’d like to confirm if the network connectivity to be provided by ICTO includes the necessary network hardware, or will the Cloud be provided only with the necessary ports. Connectivity to the public Internet will be provided by the iGovPhil Project. This means that a single (or double for redundancy) Ethernet or (Fiber Optic) cable(s) will be provided for the NextGen cloud to connect to the Internet.  For internal networking, the winning bidder should bundle switches with their proposed solution as described in the TOR.
15 We’d like to check if there is a desired Bandwidth size for the connectivity between the Data Centers? There is no desired bandwidth size for the connectivity between the Data Centers. Currently, the connection between the two is at 10 Gbps provided by the iGovPhil Project.
16 With regard to security, we request that DOST-ICTO share the information related to GovCloud’s PKI. The iGovPhil’s PNPKI implementation useS EJBCA.  EJBCA documentation can be found at their website (www.ejbca.org). For SSO, iGovPhil uses Gluu. The documentation can be found at www.gluu.org.
17 We request for information regarding the Software Defined Network (SDN) Technology to be adopted by DOST-ICTO. We need this to recommend the best suited solution aligned with ICTO’s roadmap. While different vendors may have different SDN technology implementations, the TOR has only two (2) basic requirements. (1) All switches bundled with the solution must be SDN-ready. (2) The switches must work with the cloud orchestration platform.
18 For the Proof-of-Concept required for the Contractor, we request for the expected coverage of the POC in terms of showcasing the functions and features of the cloud stack proposed. We also request for the allotted time per Contractor for the POC. The coverage of the POC is described in Supplemental Bid Bulletin No. 1.  Each contractor is given at most 4 hours to demo and answer the questions.
19 1.4.1.12, under the Capacity Monitoring, for over subscription, what is the ratio of the vCPU against the Physical cores? The ratio of vCPU against physical cores for over subscription is 4:1.
20 What should be the minimum capacity of Disk for the Backup-to-disk and backup-to-tape items in the TOR (1.4.1.43) The minimum capacity of the Backup to disk service should be 50TB. With regard to the backup to tape service, assuming no compression, using LTO6 tapes, and 50 TB target full backup, we need a minimum of 100 LTO6 tapes.
21 1.4.1.11 what are the details that comprise resources type? Examples of Resource types can be: IP addresses, disks, CPU, RAM, bandwidth, etc. Basically, we need a way to monitor how all these resources are utilized so that we can plan accordingly.
22 1.4.1.29 what’s the encompassing scope of provisioning for physical servers? Section 1.4.1.29 does not mention anything about provisioning for physical servers.  Please check if you are referring to the correct section in the TOR.
23 1.4.1.50 what is DOST-ICTO’s security governance standard policy and procedure? There is a list of policies and standards related to security found at ICTO’s website (http://dict.gov.ph/cybersecurity/)Furthermore, DOST also has an ICT usage and security policy published online. It can be found at http://www.ncc.gov.ph/files/itpolicy.pdf
24 On Business Severity Incidents, for Service Incident Resolution Time, how are the severity levels defined? A Business Severity Definition is included in the TOR. It describes the circumstances by which an incident is classified as level 1, 2 or 3, response times, and resolution times.
25 On Network Performance SLA – we’d like to check if the incidents are on the network infrastructure side or VM infrastructure part. With regard to network performance SLA, the winning bidder will only respond to issues in the VM infrastructure part.
26 1.4.1.2, in relation to ubiquitous network, does this include the portal for the VM? Yes, the self-service portal should also be ubiquitously accessible.
27 With regards to the Service Definition, may the Service Delivery Process be further clarified and expounded? Reference the ITIL Service Delivery process.
28 1.4.1.8 For the billing, will it only encompass full-blown and charge-back system type of billing mechanism? For the moment, the iGovPhil does not charge government agencies for the use of the current government cloud.  The policy not to charge on the NextGen Govcloud has not charged.  Be that as it may, we require a billing system be in place for reportorial purposes, i.e. report on how physical/virtual resources are translated into monetary terms. Whether it is a full-blown or charge-back billing mechanism has not been decided yet.
29 On items related to the Dashboard, we request that the level of access for “dashboard admin” and “dashboard user” be clarified. Section 3.3.2.2 does in fact outline the differences between what a cloud administrator and a cloud user can do in the dashboard. For a more detailed description of the two types of users, it can be provided during the implementation of the project.
30 On item 1.4.1.9, what do the terms “platform tiers”, “infrastructure platform”, and “platform resources” include? Also, is there any reference of the flow process? The GovCloud is composed of different tiers which we aptly call platform tiers. With regard to the flow process reference being requested, there is no formal process defined at this point. It is expected that the flow process will be drafted and implemented with the cooperation of DOST-ICTO and the winning bidder during project execution.
31 Also in relation to 1.4.1.9, we’d like to confirm that this section refers to infra-related resources and not the app-level resources. If you mean by app-level resources, you are referring to the client apps installed individually in each virtual machine, then Section 1.4.1.9 does not apply.
32 1.4.1.12, for Metering and Capacity monitoring. From an application workload perspective, not all application can scale. Is this an essential requirement needed to be implemented in GovCloud platform? Yes. To say that “not all applications can scale” is not entirely true.  An application may not scale if you add more virtual machine instances to it. However, it may scale up if you add enough memory, CPU or disk space to the single VM instance it is using. Legacy apps using 32 bit systems may be problematic scaling horizontally or vertically, but they are more of an exception rather than the rule.
33 1.4.1.14, can DOST-ICTO give sample of analytics, trending patterns, and matrix? The Service Analytics component derives its data from the Metering, Capacity Monitoring, and Performance Monitoring subcomponents. Due to the varied nature of data being collected, by bringing them together actually gives us potential insights on what we can plan to improve the service.  Basic information such as: How much of our resources allocated to our users are actually being used and when will actual 80% utilization take place?  Who among our users are consuming our resources and what is their impact on the overall performance of the platform? Based on the rate we are provisioning resources for the last 3-6 months, how long before we need to procure more resources?

 

34 1.4.1.15, Service Desk, we’d like to confirm that DOST users will directly contact service desk hotline and there is no integration on portal. Ideally, the service desk facility should be integrated with the portal. You may propose a separate service facility though, if the main solution does not have an integrated one. To directly answer your question, DOST users can directly contact service desk hotline and that integration with the portal is optional. At a minimum, hotline contact information should be made available through the portal.
35 1.4.1.21 we’d like to confirm if it will be necessary to update the version of the GovCloud infrastructure proposed based on the product roadmap. No. There is no mention in Section 1.4.1.21 that requires the contractor to update / upgrade the platform based on the product roadmap. It mentions about upgrading the GovCloud, but given the short contract period (1 year) and the fact that all of the equipment and software are certified to be supported by manufacturers in the next three years, these two circumstances actually minimizes the chances of the new GovCloud being upgraded within a year of its implementation.
36 1.4.1.25, we request for the classifications of downtime / real-time issues. We also request guidance as to how downtime will be conducted (i.e. scheduled, unscheduled). For classification of downtimes, please refer to the definition of Business Severity in the TOR.  Unscheduled downtimes are of course unavoidable but can be managed.  For scheduled downtimes, please refer to section 3.3.4.5.
37 1.4.1.40, Compute standards, we request for the classification of dormant VMs. What does it mean by “authenticated before they are allowed to boot”? Dormant VMs are defined as VMs having been provisioned already, potentially have been run before, but are currently in the OFF state i.e. not running.
38 1.4.1.40, we request for information as to the security policies and software policies being referred to. There is no formal DOST-ICTO security policy yet on VMs and Operating Systems.  We hope to draft these policies with the Contractor’s help during the implementation.
39 We’d like to know of DOST will provide the necessary racks on the data centers, for the equipment to be deployed. Yes, we can provide the racks.  However, we encourage you to visit our data centers, review the specifications of our racks and see if they will meet the requirements of your equipment. Otherwise, you will be bringing in your own racks to the data centers.
40 Topic: Bid Security

Reference Number:C.15.1

Page No.: 52

Clarification: We’d like to confirm that if we submit bid  security in the for of cash, manager’s check, or security bond, there will be no need for a bid securing declaration. Since we have already provided for bid security, it seems excessive to require bid securing declaration

Bid security shall be issued in favor of the Information and  Communications Technology Office (ICT Office) valid at least one hundred twenty (120) days after date of bid opening in any of the following forms:

a) Bid Securing Declaration per Annex II of the bidding documents; OR

b) Cashier’s / Manager’s Check equivalent to at least 2% of ABC issued by a Local or Universal or Local Commercial Bank.

41 Topic: Managed Services – Designing and Provisioning Operating System

Reference No.: 1.4.1.44

Page No.: 130

Clarification: We would like to further clarify what the term “visually design infrastructure services” mean.

GovCloud services are available online via the self-service portal.  To configure the services and resources that are being offered by DOST-ICTO, an agency must login to the portal and be able to setup these resources according to their requirements. In order for these resources (e.g. Physical servers, virtual machines, shared storage, virtual networks, etc.) to be easily managed by an agency, they must be “visually” conceptualized. What this means is that resources can be ideally dragged and dropped from a catalogue and connected to each other, clicked on to be configured, and moved around for easy understanding of the network setup.  This is what the “visually design infrastructure services” means.
42 Topic: Managed Services: Designing & Provisioning Operating System

Reference No.: 1.4.1.45

Page No.: 130

Clarification: We would like to clarify if it is necessary to provision a physical or a bare metal server.

No. All computes are understood to be virtual i.e. instantiated on virtual machines. However, the capability to instantiate on a physical machine should be there.
43 Topic: Managed Services

Reference No.: 1.4.1.44

Page: 130

Clarification: With regard to the term “Managed  Services”, can DOST-ICTO further elaborate the scope and implications of its included subtopics (i.e. Designing & Provisioning OS, Central & Unified Management for Physical & Virtual Machines, Capacity Monitoring, etc.)?

From our perspective, the section 1..4.1.44 (Managed Services) is as detailed as it can get.  Each item in the section is actually a feature that your proposed solution should be able to meet. If there are certain features mentioned in the section that are unclear, we would like to you to point it out.
44 Topic: Managed Services – Central and Unified Management for Physical and Virtual Machines

Reference No.: 1.4.1.45

Page No.: 131

Clarification: What does it mean by the term “predictive hardware failure?’

Certain technologies which vary from manufacturer to manufacturer, allow server components to be monitored so that the server will be able to detect if something is wrong and adjust/rectify the error or notify the server admin. This feature of the hardware is what we call as predictive failure analysis and should be compatible with the management software. When a predictive hardware failure happens, the management software should be able to detect it and adjust accordingly.
45 Topic: N/A

Reference No.: N/A

Page No.: N/A

Clarification: As mentioned on several items of the TOR, what does it mean to have “physical servers” (i.e. physical provisioning, physical isolation, etc.)? Our assumption is, sine this is a cloud deployment project, virtual environment will be offered.

The answer to this question is related to question #2.  It is true that a virtual environment is what is being required in the GovCloud project. However, as mentioned in the answer to question #2, the capability to instantiate a compute on a physical machine is required.  In general, this requirement to be able to provision for physical resources applies to all the components not just compute.  For example, we want the capability to provision a physical network switch without configuring it directly but accessible via the management software. Provisioning for network storage so that it can be available to a physical server previously instantiated via the management software is another example. The GovCloud will only provision for virtual resources but the capability to provision for physical resources should be there as well.
46 Topic: Disaster Recovery Capability

Reference No.:1.4.1.48

Page No.: 133

Clarification: We’d like to request DOST-ICTO to further elaborate the scope of Disaster Recovery and how will it be performed.

The scope of disaster recovery is the transitioning of workload from the main site to a secondary site in case of a large scale disaster.  Please note though that by our own convention, we consider both sites as Active.  Some applications may be deployed to the Makati NGDC and others may be deployed to the Quezon City NGDC. There is no preference or defined convention as to how applications will be deployed to these data centers. When critical applications are affected in one data center, they can be transferred to the other data center. Please note that we put great emphasis on critical applications only as this will only apply to a small subset of the workload being run in one data center. Not all the applications will have to be recovered in the other data center.  Further details about the disaster recovery capability are already found in section 1.4.1.48.
47 Topic: Security Governance

Reference No.: 1.4.1.50

Page No.: 135

Clarification: We’d like to ask for DOST’s architecture design.

Please note that the architecture design should come from the contractor as mentioned in section 1.4.1.50.
48 Topic: Security Governance

Reference No.: 1.4.1.50

Page No.: 135

Clarification: We’d like to further clarify the definition of the phrase “intranet agency network”.

The intranet agency network refers to the agency’s LAN. Some applications in the cloud need only to be accessible to an agency’s LAN. There must be a way to seamlessly expose the application that is in the cloud to the agency’s LAN while being secure and managed through the management software.

 

49 Topic: Portal Authentication

Reference No.: 1.4.1.53

Page No.: 136

Clarification: We’d like to ask for both PKI and SDK documentation of DOST GovCloud for portal and infra integration.

Please see Item No. 16.
50 Topic: Centralized Audit Trail

Reference No.: 1.4.1.58

Page No.: 138

Clarification: With regard centralized audit trail, we’d like to know further the expected outcome DOST-ICTO is expecting for the audit trail.

The outcome of the centralized audit trail is to have a system whereby DOST-ICTO will be able to determine who did what, when, how, and possibly why it was done.
51 Topic: Training and Awareness

Reference No.: 1.4.1.59

Page No.: 139

Clarification: We’d like to DOST to further include the number of participants involved in the trainings mentioned in the TOR.

The estimated number of training participants will be at most 200. They can be trained in batches and scheduled for the whole year of the implementation based on the training plan submitted by the contractor.
52 Topic: Disaster Recovery Capability

Reference No.: 1.4.1.48

Page No.: 133

Clarification: We’d like DOST to further define the required Recovery Time Objective (RTO) and Recovery Point Objective (RPO) incases of DR activities.

The RTO should be 4 hours.  The RPO should be 30 mins.  Once the, main site goes up, it should take at most 8 hours to revert to the main site.
53 Topic: Evaluation Criteria of Personnel Assigned to the Project

Reference No.: 3.1.3

Page No.: 153

Clarification: We’d like to confirm that for the Resident Project Manager, the equivalent score for Relevant Experience is 0.

This is an oversight on our part. Here’s the corrected distribution of points for the Resident Project Manager evaluation criteria. Qualifications: 3, Overall Experience: 2, Relevant Experience: 3

 

54 Topic: N/A

Reference No.: N/A

Page No.: N/A

Clarification: We’d like DOST to further provide the details regarding the current ISP and bandwidth being used (including point to point connectivity of the 2 data centers, internet connectivity, etc.)

Current Internet bandwidth is 1 Gbps. Projected bandwidth at the end of March is 10 Gbps.  Bandwidth between the two data centers is 10 Gbps.

 

 

All terms, conditions and instructions to bidders specified in the Bidding Documents inconsistent with this Bid Bulletin are hereby superseded and modified accordingly.

To download this bidding document, click on the link below:

Supplemental Bid Bulletin No. 4 – Consultancy Services for the Design, Build, and Operate of a Complete Cloud Solution for Philippine Government Agencies

For information and guidance of all concerned.

Issued this 26th day of February 2016.

(SGD) DENIS F. VILLORENTE
Chairman, BAC4IGOV